DonorsChoose

DonorsChoose.org receives hundreds of thousands of project proposals each year for classroom projects in need of funding. Right now, a large number of volunteers is needed to manually screen each submission before it's approved to be posted on the DonorsChoose.org website.

Next year, DonorsChoose.org expects to receive close to 500,000 project proposals. As a result, there are three main problems they need to solve:

• How to scale current manual processes and resources to screen 500,000 projects so that they can be posted as quickly and as efficiently as possible
• How to increase the consistency of project vetting across different volunteers to improve the experience for teachers
• How to focus volunteer time on the applications that need the most assistance

The goal of the competition is to predict whether or not a DonorsChoose.org project proposal submitted by a teacher will be approved, using the text of project descriptions as well as additional metadata about the project, teacher, and school. DonorsChoose.org can then use this information to identify projects most likely to need further review before approval.

About the DonorsChoose Data Set

The train.csv data set provided by DonorsChoose contains the following features:

Feature	Description
project_id	A unique identifier for the proposed project. Example: p036502
project_title	Title of the project. Examples: Art Will Make You Happy! First Grade Fun
project_grade_category	Grade level of students for which the project is targeted. One of the following enumerated values: Grades PreK-2 Grades 3-5 Grades 6-8 Grades 9-12
project_subject_categories	One or more (comma-separated) subject categories for the project from the following enumerated list of values: Applied Learning Care & Hunger Health & Sports History & Civics Literacy & Language Math & Science Music & The Arts Special Needs Warmth Examples: Music & The Arts Literacy & Language, Math & Science
school_state	State where school is located (Two-letter U.S. postal code). Example: WY
project_subject_subcategories	One or more (comma-separated) subject subcategories for the project. Examples: Literacy Literature & Writing, Social Sciences
project_resource_summary	An explanation of the resources needed for the project. Example: My students need hands on literacy materials to manage sensory needs!
project_essay_1	First application essay*
project_essay_2	Second application essay*
project_essay_3	Third application essay*
project_essay_4	Fourth application essay*
project_submitted_datetime	Datetime when project application was submitted. Example: 2016-04-28 12:43:56.245
teacher_id	A unique identifier for the teacher of the proposed project. Example: bdf8baa8fedef6bfeec7ae4ff1c15c56
teacher_prefix	Teacher's title. One of the following enumerated values: nan Dr. Mr. Mrs. Ms. Teacher.
teacher_number_of_previously_posted_projects	Number of project applications previously submitted by the same teacher. Example: 2

^* See the section Notes on the Essay Data for more details about these features.

Additionally, the resources.csv data set provides more data about the resources required for each project. Each line in this file represents a resource required by a project:

Feature	Description
id	A project_id value from the train.csv file. Example: p036502
description	Desciption of the resource. Example: Tenor Saxophone Reeds, Box of 25
quantity	Quantity of the resource required. Example: 3
price	Price of the resource required. Example: 9.95

Note: Many projects require multiple resources. The id value corresponds to a project_id in train.csv, so you use it as a key to retrieve all resources needed for a project:

The data set contains the following label (the value you will attempt to predict):

Label	Description
project_is_approved	A binary flag indicating whether DonorsChoose approved the project. A value of 0 indicates the project was not approved, and a value of 1 indicates the project was approved.

Notes on the Essay Data

Prior to May 17, 2016, the prompts for the essays were as follows:
• __project_essay_1:__ "Introduce us to your classroom"
• __project_essay_2:__ "Tell us more about your students"
• __project_essay_3:__ "Describe how your students will use the materials you're requesting"
• __project_essay_3:__ "Close by sharing why your project will make a difference"

Starting on May 17, 2016, the number of essays was reduced from 4 to 2, and the prompts for the first 2 essays were changed to the following:

• __project_essay_1:__ "Describe your students: What makes your students special? Specific details about their background, your neighborhood, and your school are all helpful."
• __project_essay_2:__ "About your project: How will these materials make a difference in your students' learning and improve their school lives?"

For all projects with project_submitted_datetime of 2016-05-17 and later, the values of project_essay_3 and project_essay_4 will be NaN.

%matplotlib inline
import warnings
warnings.filterwarnings("ignore")

import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer

import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer

from gensim.models import Word2Vec
from gensim.models import KeyedVectors
import pickle

from tqdm import tqdm
import os

from chart_studio import plotly
import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter

1.1 Reading Data

project_data = pd.read_csv('../train_data.csv')
resource_data = pd.read_csv('../resources.csv')

print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)

Number of data points in train data (109248, 17)
--------------------------------------------------
The attributes of data : ['Unnamed: 0' 'id' 'teacher_id' 'teacher_prefix' 'school_state'
 'project_submitted_datetime' 'project_grade_category'
 'project_subject_categories' 'project_subject_subcategories'
 'project_title' 'project_essay_1' 'project_essay_2' 'project_essay_3'
 'project_essay_4' 'project_resource_summary'
 'teacher_number_of_previously_posted_projects' 'project_is_approved']

print("Number of data points in train data", resource_data.shape)
print(resource_data.columns.values)
resource_data.head(2)

Number of data points in train data (1541272, 4)
['id' 'description' 'quantity' 'price']

	id	description	quantity	price
0	p233245	LC652 - Lakeshore Double-Space Mobile Drying Rack	1	149.00
1	p069063	Bouncy Bands for Desks (Blue support pipes)	3	14.95

1.2 Preprocessing Categorical Data

1.2.1 preprocessing project_subject_categories

catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039

# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
    temp = ""
    # consider we have text like this "Math & Science, Warmth, Care & Hunger"
    for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
        if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
            j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
        j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
        temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
        temp = temp.replace('&','_') # we are replacing the & value into 
    cat_list.append(temp.strip())
    
project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)

from collections import Counter
my_counter = Counter()
for word in project_data['clean_categories'].values:
    my_counter.update(word.split())

cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))

sorted_cat_dict.keys()

dict_keys(['Warmth', 'Care_Hunger', 'History_Civics', 'Music_Arts', 'AppliedLearning', 'SpecialNeeds', 'Health_Sports', 'Math_Science', 'Literacy_Language'])

1.2.2 preprocessing of project_subject_subcategories

sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039

# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python

sub_cat_list = []
for i in sub_catogories:
    temp = ""
    # consider we have text like this "Math & Science, Warmth, Care & Hunger"
    for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
        if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
            j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
        j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
        temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
        temp = temp.replace('&','_')
    sub_cat_list.append(temp.strip())

project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)

# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
my_counter = Counter()
for word in project_data['clean_subcategories'].values:
    my_counter.update(word.split())
    
sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))

sorted_sub_cat_dict.keys()

dict_keys(['Economics', 'CommunityService', 'FinancialLiteracy', 'ParentInvolvement', 'Extracurricular', 'Civics_Government', 'ForeignLanguages', 'NutritionEducation', 'Warmth', 'Care_Hunger', 'SocialSciences', 'PerformingArts', 'CharacterEducation', 'TeamSports', 'Other', 'College_CareerPrep', 'Music', 'History_Geography', 'Health_LifeScience', 'EarlyDevelopment', 'ESL', 'Gym_Fitness', 'EnvironmentalScience', 'VisualArts', 'Health_Wellness', 'AppliedSciences', 'SpecialNeeds', 'Literature_Writing', 'Mathematics', 'Literacy'])

1.2.3 preprocessing of School State

project_data['school_state'].unique()

array(['IN', 'FL', 'AZ', 'KY', 'TX', 'CT', 'GA', 'SC', 'NC', 'CA', 'NY',
       'OK', 'MA', 'NV', 'OH', 'PA', 'AL', 'LA', 'VA', 'AR', 'WA', 'WV',
       'ID', 'TN', 'MS', 'CO', 'UT', 'IL', 'MI', 'HI', 'IA', 'RI', 'NJ',
       'MO', 'DE', 'MN', 'ME', 'WY', 'ND', 'OR', 'AK', 'MD', 'WI', 'SD',
       'NE', 'NM', 'DC', 'KS', 'MT', 'NH', 'VT'], dtype=object)

project_data['school_state'][project_data['school_state'].isnull()==True]

Series([], Name: school_state, dtype: object)

# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
my_counter = Counter()
for word in project_data['school_state'].values:
    my_counter.update(word.split())
    
school_state_dict = dict(my_counter)
sorted_school_state_dict = dict(sorted(school_state_dict.items(), key=lambda kv: kv[1]))

sorted_school_state_dict.keys()

dict_keys(['VT', 'WY', 'ND', 'MT', 'RI', 'SD', 'NE', 'DE', 'AK', 'NH', 'WV', 'ME', 'HI', 'DC', 'NM', 'KS', 'IA', 'ID', 'AR', 'CO', 'MN', 'OR', 'KY', 'MS', 'NV', 'MD', 'CT', 'TN', 'UT', 'AL', 'WI', 'VA', 'AZ', 'NJ', 'OK', 'WA', 'MA', 'LA', 'OH', 'MO', 'IN', 'PA', 'MI', 'SC', 'GA', 'IL', 'NC', 'FL', 'NY', 'TX', 'CA'])

1.2.4 preprocessing of Teacher Prefix

project_data.groupby(['teacher_prefix'])['teacher_prefix'].count()

teacher_prefix
Dr.           13
Mr.        10648
Mrs.       57269
Ms.        38955
Teacher     2360
Name: teacher_prefix, dtype: int64

project_data['teacher_prefix'][project_data['teacher_prefix'].isnull()==True]

7820     NaN
30368    NaN
57654    NaN
Name: teacher_prefix, dtype: object

project_data['teacher_prefix'].fillna(project_data['teacher_prefix'].mode()[0],inplace=True)

project_data['teacher_prefix'][project_data['teacher_prefix'].isnull()==True]

Series([], Name: teacher_prefix, dtype: object)

project_data['teacher_prefix'].unique()

array(['Mrs.', 'Mr.', 'Ms.', 'Teacher', 'Dr.'], dtype=object)

teacher_prefix = list(project_data['teacher_prefix'].values)

teacher_prefix_list = []
for i in teacher_prefix:
    temp = ""
    temp = i.split('.')
    temp = i.replace('.','')   
    teacher_prefix_list.append(temp)

project_data['clean_teacher_prefix'] = teacher_prefix_list
project_data.drop(['teacher_prefix'], axis=1, inplace=True)

# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
my_counter = Counter()
for word in project_data['clean_teacher_prefix'].values:
    my_counter.update(word.split())
    
teacher_prefix_dict = dict(my_counter)
sorted_teacher_prefix_dict = dict(sorted(teacher_prefix_dict.items(), key=lambda kv: kv[1]))

sorted_teacher_prefix_dict.keys()

dict_keys(['Dr', 'Teacher', 'Mr', 'Ms', 'Mrs'])

project_data.groupby(['clean_teacher_prefix'])['clean_teacher_prefix'].count()

clean_teacher_prefix
Dr            13
Mr         10648
Mrs        57272
Ms         38955
Teacher     2360
Name: clean_teacher_prefix, dtype: int64

1.2.5 preprocessing of Project Grade Category

project_data.groupby(['project_grade_category'])['project_grade_category'].count()

project_grade_category
Grades 3-5       37137
Grades 6-8       16923
Grades 9-12      10963
Grades PreK-2    44225
Name: project_grade_category, dtype: int64

project_data['project_grade_category'][project_data['project_grade_category'].isnull()==True]

Series([], Name: project_grade_category, dtype: object)

project_grade_category = list(project_data['project_grade_category'].values)

project_grade_category_list = []
for i in project_grade_category:
    temp = ""
    temp = i.split(' ')
    temp = i.replace('Grades ','')   
    project_grade_category_list.append(temp)

project_data['clean_project_grade_category'] = project_grade_category_list
project_data.drop(['project_grade_category'], axis=1, inplace=True)

# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
my_counter = Counter()
for word in project_data['clean_project_grade_category'].values:
    my_counter.update(word.split())
    
project_grade_category_dict = dict(my_counter)
sorted_project_grade_category_dict = dict(sorted(project_grade_category_dict.items(), key=lambda kv: kv[1]))

sorted_project_grade_category_dict.keys()

dict_keys(['9-12', '6-8', '3-5', 'PreK-2'])

project_data.groupby(['clean_project_grade_category'])['clean_project_grade_category'].count()

clean_project_grade_category
3-5       37137
6-8       16923
9-12      10963
PreK-2    44225
Name: clean_project_grade_category, dtype: int64

1.3 Text preprocessing

# merge two column text dataframe: 
project_data["essay"] = project_data["project_essay_1"].map(str) +\
                        project_data["project_essay_2"].map(str) + \
                        project_data["project_essay_3"].map(str) + \
                        project_data["project_essay_4"].map(str)

project_data.head(2)

	Unnamed: 0	id	teacher_id	school_state	project_submitted_datetime	project_title	project_essay_1	project_essay_2	project_essay_3	project_essay_4	project_resource_summary	teacher_number_of_previously_posted_projects	project_is_approved	clean_categories	clean_subcategories	clean_teacher_prefix	clean_project_grade_category	essay
0	160221	p253737	c90749f5d961ff158d4b4d1e7dc665fc	IN	2016-12-05 13:43:57	Educational Support for English Learners at Home	My students are English learners that are work...	\"The limits of your language are the limits o...	NaN	NaN	My students need opportunities to practice beg...	0	0	Literacy_Language	ESL Literacy	Mrs	PreK-2	My students are English learners that are work...
1	140945	p258326	897464ce9ddc600bced1151f324dd63a	FL	2016-10-25 09:22:10	Wanted: Projector for Hungry Learners	Our students arrive to our school eager to lea...	The projector we need for our school is very c...	NaN	NaN	My students need a projector to help with view...	7	1	History_Civics Health_Sports	Civics_Government TeamSports	Mr	6-8	Our students arrive to our school eager to lea...

#### 1.4.2.3 Using Pretrained Models: TFIDF weighted W2V

# printing some random reviews
print(project_data['essay'].values[0])
print("="*50)
print(project_data['essay'].values[150])
print("="*50)
print(project_data['essay'].values[1000])
print("="*50)
print(project_data['essay'].values[20000])
print("="*50)
print(project_data['essay'].values[99999])
print("="*50)

My students are English learners that are working on English as their second or third languages. We are a melting pot of refugees, immigrants, and native-born Americans bringing the gift of language to our school. \r\n\r\n We have over 24 languages represented in our English Learner program with students at every level of mastery.  We also have over 40 countries represented with the families within our school.  Each student brings a wealth of knowledge and experiences to us that open our eyes to new cultures, beliefs, and respect.\"The limits of your language are the limits of your world.\"-Ludwig Wittgenstein  Our English learner's have a strong support system at home that begs for more resources.  Many times our parents are learning to read and speak English along side of their children.  Sometimes this creates barriers for parents to be able to help their child learn phonetics, letter recognition, and other reading skills.\r\n\r\nBy providing these dvd's and players, students are able to continue their mastery of the English language even if no one at home is able to assist.  All families with students within the Level 1 proficiency status, will be a offered to be a part of this program.  These educational videos will be specially chosen by the English Learner Teacher and will be sent home regularly to watch.  The videos are to help the child develop early reading skills.\r\n\r\nParents that do not have access to a dvd player will have the opportunity to check out a dvd player to use for the year.  The plan is to use these videos and educational dvd's for the years to come for other EL students.\r\nnannan
==================================================
The 51 fifth grade students that will cycle through my classroom this year all love learning, at least most of the time. At our school, 97.3% of the students receive free or reduced price lunch. Of the 560 students, 97.3% are minority students. \r\nThe school has a vibrant community that loves to get together and celebrate. Around Halloween there is a whole school parade to show off the beautiful costumes that students wear. On Cinco de Mayo we put on a big festival with crafts made by the students, dances, and games. At the end of the year the school hosts a carnival to celebrate the hard work put in during the school year, with a dunk tank being the most popular activity.My students will use these five brightly colored Hokki stools in place of regular, stationary, 4-legged chairs. As I will only have a total of ten in the classroom and not enough for each student to have an individual one, they will be used in a variety of ways. During independent reading time they will be used as special chairs students will each use on occasion. I will utilize them in place of chairs at my small group tables during math and reading times. The rest of the day they will be used by the students who need the highest amount of movement in their life in order to stay focused on school.\r\n\r\nWhenever asked what the classroom is missing, my students always say more Hokki Stools. They can't get their fill of the 5 stools we already have. When the students are sitting in group with me on the Hokki Stools, they are always moving, but at the same time doing their work. Anytime the students get to pick where they can sit, the Hokki Stools are the first to be taken. There are always students who head over to the kidney table to get one of the stools who are disappointed as there are not enough of them. \r\n\r\nWe ask a lot of students to sit for 7 hours a day. The Hokki stools will be a compromise that allow my students to do desk work and move at the same time. These stools will help students to meet their 60 minutes a day of movement by allowing them to activate their core muscles for balance while they sit. For many of my students, these chairs will take away the barrier that exists in schools for a child who can't sit still.nannan
==================================================
How do you remember your days of school? Was it in a sterile environment with plain walls, rows of desks, and a teacher in front of the room? A typical day in our room is nothing like that. I work hard to create a warm inviting themed room for my students look forward to coming to each day.\r\n\r\nMy class is made up of 28 wonderfully unique boys and girls of mixed races in Arkansas.\r\nThey attend a Title I school, which means there is a high enough percentage of free and reduced-price lunch to qualify. Our school is an \"open classroom\" concept, which is very unique as there are no walls separating the classrooms. These 9 and 10 year-old students are very eager learners; they are like sponges, absorbing all the information and experiences and keep on wanting more.With these resources such as the comfy red throw pillows and the whimsical nautical hanging decor and the blue fish nets, I will be able to help create the mood in our classroom setting to be one of a themed nautical environment. Creating a classroom environment is very important in the success in each and every child's education. The nautical photo props will be used with each child as they step foot into our classroom for the first time on Meet the Teacher evening. I'll take pictures of each child with them, have them developed, and then hung in our classroom ready for their first day of 4th grade.  This kind gesture will set the tone before even the first day of school! The nautical thank you cards will be used throughout the year by the students as they create thank you cards to their team groups.\r\n\r\nYour generous donations will help me to help make our classroom a fun, inviting, learning environment from day one.\r\n\r\nIt costs lost of money out of my own pocket on resources to get our classroom ready. Please consider helping with this project to make our new school year a very successful one. Thank you!nannan
==================================================
My kindergarten students have varied disabilities ranging from speech and language delays, cognitive delays, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest working past their limitations. \r\n\r\nThe materials we have are the ones I seek out for my students. I teach in a Title I school where most of the students receive free or reduced price lunch.  Despite their disabilities and limitations, my students love coming to school and come eager to learn and explore.Have you ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting? This is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble chairs are the answer and I love then because they develop their core, which enhances gross motor and in Turn fine motor skills. \r\nThey also want to learn through games, my kids don't want to sit and do worksheets. They want to learn to count by jumping and playing. Physical engagement is the key to our success. The number toss and color and shape mats can make that happen. My students will forget they are doing work and just have the fun a 6 year old deserves.nannan
==================================================
The mediocre teacher tells. The good teacher explains. The superior teacher demonstrates. The great teacher inspires. -William A. Ward\r\n\r\nMy school has 803 students which is makeup is 97.6% African-American, making up the largest segment of the student body. A typical school in Dallas is made up of 23.2% African-American students. Most of the students are on free or reduced lunch. We aren't receiving doctors, lawyers, or engineers children from rich backgrounds or neighborhoods. As an educator I am inspiring minds of young children and we focus not only on academics but one smart, effective, efficient, and disciplined students with good character.In our classroom we can utilize the Bluetooth for swift transitions during class. I use a speaker which doesn't amplify the sound enough to receive the message. Due to the volume of my speaker my students can't hear videos or books clearly and it isn't making the lessons as meaningful. But with the bluetooth speaker my students will be able to hear and I can stop, pause and replay it at any time.\r\nThe cart will allow me to have more room for storage of things that are needed for the day and has an extra part to it I can use.  The table top chart has all of the letter, words and pictures for students to learn about different letters and it is more accessible.nannan
==================================================

# https://stackoverflow.com/a/47091490/4084039
import re

def decontracted(phrase):
    # specific
    phrase = re.sub(r"won't", "will not", phrase)
    phrase = re.sub(r"can\'t", "can not", phrase)

    # general
    phrase = re.sub(r"n\'t", " not", phrase)
    phrase = re.sub(r"\'re", " are", phrase)
    phrase = re.sub(r"\'s", " is", phrase)
    phrase = re.sub(r"\'d", " would", phrase)
    phrase = re.sub(r"\'ll", " will", phrase)
    phrase = re.sub(r"\'t", " not", phrase)
    phrase = re.sub(r"\'ve", " have", phrase)
    phrase = re.sub(r"\'m", " am", phrase)
    return phrase

sent = decontracted(project_data['essay'].values[20000])
print(sent)
print("="*50)

My kindergarten students have varied disabilities ranging from speech and language delays, cognitive delays, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest working past their limitations. \r\n\r\nThe materials we have are the ones I seek out for my students. I teach in a Title I school where most of the students receive free or reduced price lunch.  Despite their disabilities and limitations, my students love coming to school and come eager to learn and explore.Have you ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting? This is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble chairs are the answer and I love then because they develop their core, which enhances gross motor and in Turn fine motor skills. \r\nThey also want to learn through games, my kids do not want to sit and do worksheets. They want to learn to count by jumping and playing. Physical engagement is the key to our success. The number toss and color and shape mats can make that happen. My students will forget they are doing work and just have the fun a 6 year old deserves.nannan
==================================================

# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)

My kindergarten students have varied disabilities ranging from speech and language delays, cognitive delays, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest working past their limitations.     The materials we have are the ones I seek out for my students. I teach in a Title I school where most of the students receive free or reduced price lunch.  Despite their disabilities and limitations, my students love coming to school and come eager to learn and explore.Have you ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting? This is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble chairs are the answer and I love then because they develop their core, which enhances gross motor and in Turn fine motor skills.   They also want to learn through games, my kids do not want to sit and do worksheets. They want to learn to count by jumping and playing. Physical engagement is the key to our success. The number toss and color and shape mats can make that happen. My students will forget they are doing work and just have the fun a 6 year old deserves.nannan

#remove spacial character: https://stackoverflow.com/a/5843547/4084039
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)

My kindergarten students have varied disabilities ranging from speech and language delays cognitive delays gross fine motor delays to autism They are eager beavers and always strive to work their hardest working past their limitations The materials we have are the ones I seek out for my students I teach in a Title I school where most of the students receive free or reduced price lunch Despite their disabilities and limitations my students love coming to school and come eager to learn and explore Have you ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting This is how my kids feel all the time The want to be able to move as they learn or so they say Wobble chairs are the answer and I love then because they develop their core which enhances gross motor and in Turn fine motor skills They also want to learn through games my kids do not want to sit and do worksheets They want to learn to count by jumping and playing Physical engagement is the key to our success The number toss and color and shape mats can make that happen My students will forget they are doing work and just have the fun a 6 year old deserves nannan

# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
            "you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
            'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
            'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
            'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
            'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
            'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
            'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
            'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
            'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
            's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
            've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
            "hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
            "mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
            'won', "won't", 'wouldn', "wouldn't"]

# Combining all the above stundents 
from tqdm import tqdm
preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['essay'].values):
    sent = decontracted(sentance)
    sent = sent.replace('\\r', ' ')
    sent = sent.replace('\\"', ' ')
    sent = sent.replace('\\n', ' ')
    sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
    # https://gist.github.com/sebleier/554280
    sent = ' '.join(e for e in sent.split() if e not in stopwords)
    preprocessed_essays.append(sent.lower().strip())

100%|██████████| 109248/109248 [02:41<00:00, 677.11it/s]

# after preprocesing
preprocessed_essays[20000]

'my kindergarten students varied disabilities ranging speech language delays cognitive delays gross fine motor delays autism they eager beavers always strive work hardest working past limitations the materials ones i seek students i teach title i school students receive free reduced price lunch despite disabilities limitations students love coming school come eager learn explore have ever felt like ants pants needed groove move meeting this kids feel time the want able move learn say wobble chairs answer i love develop core enhances gross motor turn fine motor skills they also want learn games kids not want sit worksheets they want learn count jumping playing physical engagement key success the number toss color shape mats make happen my students forget work fun 6 year old deserves nannan'

project_data['preprocessed_essays'] = preprocessed_essays
project_data.drop(['essay'], axis=1, inplace=True)

1.4 Preprocessing of `project_title`

# similarly you can preprocess the titles also

project_data['project_title'][2000:2010]

2000                    Steady Stools for Active Learning
2001                                   Classroom Supplies
2002    Kindergarten Students Deserve Quality  Books a...
2003                                Listen to Understand!
2004                             iPads to iGnite Learning
2005                                 Tablets For Learning
2006                                             Go P.E.!
2007                                 Making Learning Fun!
2008    Empowerment Through Silk Screen Designed Tee S...
2009                                 Let's Play Together!
Name: project_title, dtype: object

# Combining all the above statemennts 
from tqdm import tqdm
preprocessed_titles = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['project_title'].values):
    sent = decontracted(sentance)
    sent = sent.replace('\\r', ' ')
    sent = sent.replace('\\"', ' ')
    sent = sent.replace('\\n', ' ')
    sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
    # https://gist.github.com/sebleier/554280
    sent = ' '.join(e for e in sent.split() if e not in stopwords)
    preprocessed_titles.append(sent.lower().strip())

100%|██████████| 109248/109248 [00:07<00:00, 14630.67it/s]

preprocessed_titles[2000:2010]

['steady stools active learning',
 'classroom supplies',
 'kindergarten students deserve quality books vibrant rug',
 'listen understand',
 'ipads ignite learning',
 'tablets for learning',
 'go p e',
 'making learning fun',
 'empowerment through silk screen designed tee shirts',
 'let play together']

project_data['preprocessed_titles'] = preprocessed_titles
project_data.drop(['project_title'], axis=1, inplace=True)

1.5 Preparing data for models

project_data.columns

Index(['Unnamed: 0', 'id', 'teacher_id', 'school_state',
       'project_submitted_datetime', 'project_essay_1', 'project_essay_2',
       'project_essay_3', 'project_essay_4', 'project_resource_summary',
       'teacher_number_of_previously_posted_projects', 'project_is_approved',
       'clean_categories', 'clean_subcategories', 'clean_teacher_prefix',
       'clean_project_grade_category', 'price', 'quantity',
       'preprocessed_essays', 'preprocessed_titles'],
      dtype='object')

we are going to consider

   - school_state : categorical data
   - clean_categories : categorical data
   - clean_subcategories : categorical data
   - project_grade_category : categorical data
   - teacher_prefix : categorical data

   - project_title : text data
   - text : text data
   - project_resource_summary: text data (optinal)

   - quantity : numerical 
   - teacher_number_of_previously_posted_projects : numerical
   - price : numerical

1.5.1 Vectorizing Categorical data

• https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/handling-categorical-and-numerical-features/

# we use count vectorizer to convert the values into one from sklearn.feature_extraction.text import CountVectorizer vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True) categories_one_hot = vectorizer.fit_transform(project_data['clean_categories'].values) print(vectorizer.get_feature_names()) print("Shape of matrix after one hot encodig ",categories_one_hot.shape)# we use count vectorizer to convert the values into one vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True) sub_categories_one_hot = vectorizer.fit_transform(project_data['clean_subcategories'].values) print(vectorizer.get_feature_names()) print("Shape of matrix after one hot encodig ",sub_categories_one_hot.shape)# you can do the similar thing with state, teacher_prefix and project_grade_category also

1.5.2 Vectorizing Text data

1.5.2.1 Bag of words

# We are considering only the words which appeared in at least 10 documents(rows or projects). vectorizer = CountVectorizer(min_df=10) text_bow = vectorizer.fit_transform(preprocessed_essays) print("Shape of matrix after one hot encodig ",text_bow.shape)# you can vectorize the title also # before you vectorize the title make sure you preprocess it

1.5.2.2 TFIDF vectorizer

from sklearn.feature_extraction.text import TfidfVectorizer vectorizer = TfidfVectorizer(min_df=10) text_tfidf = vectorizer.fit_transform(preprocessed_essays) print("Shape of matrix after one hot encodig ",text_tfidf.shape)

1.5.2.3 Using Pretrained Models: Avg W2V

# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039 def loadGloveModel(gloveFile): print ("Loading Glove Model") f = open(gloveFile,'r', encoding="utf8") model = {} for line in tqdm(f): splitLine = line.split() word = splitLine[0] embedding = np.array([float(val) for val in splitLine[1:]]) model[word] = embedding print ("Done.",len(model)," words loaded!") return model model = loadGloveModel('glove.42B.300d.txt') # ============================ Output: Loading Glove Model 1917495it [06:32, 4879.69it/s] Done. 1917495 words loaded! # ============================ words = [] for i in preproced_texts: words.extend(i.split(' ')) for i in preproced_titles: words.extend(i.split(' ')) print("all the words in the coupus", len(words)) words = set(words) print("the unique words in the coupus", len(words)) inter_words = set(model.keys()).intersection(words) print("The number of words that are present in both glove vectors and our coupus", \ len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)") words_courpus = {} words_glove = set(model.keys()) for i in words: if i in words_glove: words_courpus[i] = model[i] print("word 2 vec length", len(words_courpus)) # stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/ import pickle with open('glove_vectors', 'wb') as f: pickle.dump(words_courpus, f)# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/ # make sure you have the glove_vectors file with open('glove_vectors', 'rb') as f: model = pickle.load(f) glove_words = set(model.keys()) # average Word2Vec # compute average word2vec for each review. avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list for sentence in tqdm(preprocessed_essays): # for each review/sentence vector = np.zeros(300) # as word vectors are of zero length cnt_words =0; # num of words with a valid vector in the sentence/review for word in sentence.split(): # for each word in a review/sentence if word in glove_words: vector += model[word] cnt_words += 1 if cnt_words != 0: vector /= cnt_words avg_w2v_vectors.append(vector) print(len(avg_w2v_vectors)) print(len(avg_w2v_vectors[0]))

1.5.2.3 Using Pretrained Models: TFIDF weighted W2V

# S = ["abc def pqr", "def def def abc", "pqr pqr def"] tfidf_model = TfidfVectorizer() tfidf_model.fit(preprocessed_essays) # we are converting a dictionary with word as a key, and the idf as a value dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_))) tfidf_words = set(tfidf_model.get_feature_names())# average Word2Vec # compute average word2vec for each review. tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list for sentence in tqdm(preprocessed_essays): # for each review/sentence vector = np.zeros(300) # as word vectors are of zero length tf_idf_weight =0; # num of words with a valid vector in the sentence/review for word in sentence.split(): # for each word in a review/sentence if (word in glove_words) and (word in tfidf_words): vec = model[word] # getting the vector for each word # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split()))) tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word vector += (vec * tf_idf) # calculating tfidf weighted w2v tf_idf_weight += tf_idf if tf_idf_weight != 0: vector /= tf_idf_weight tfidf_w2v_vectors.append(vector) print(len(tfidf_w2v_vectors)) print(len(tfidf_w2v_vectors[0]))# Similarly you can vectorize for title also

1.5.3 Vectorizing Numerical features

price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index() project_data = pd.merge(project_data, price_data, on='id', how='left')# check this one: https://www.youtube.com/watch?v=0HOqOcln3Z4&t=530s # standardization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html from sklearn.preprocessing import StandardScaler # price_standardized = standardScalar.fit(project_data['price'].values) # this will rise the error # ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ]. # Reshape your data either using array.reshape(-1, 1) price_scalar = StandardScaler() price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}") # Now standardize the data with above maen and variance. price_standardized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))price_standardized

1.5.4 Merging all the above features

• we need to merge all the numerical vectors i.e catogorical, text, numerical vectors

print(categories_one_hot.shape) print(sub_categories_one_hot.shape) print(text_bow.shape) print(price_standardized.shape)# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039 from scipy.sparse import hstack # with the same hstack function we are concatinating a sparse matrix and a dense matirx :) X = hstack((categories_one_hot, sub_categories_one_hot, text_bow, price_standardized)) X.shape

1.6 Merging Numerical data in Resources to project_data

price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
project_data = pd.merge(project_data, price_data, on='id', how='left')

Computing Sentiment Scores

import nltk
from nltk.sentiment.vader import SentimentIntensityAnalyzer

# import nltk
# nltk.download('vader_lexicon')

sid = SentimentIntensityAnalyzer()

for_sentiment = 'a person is a person no matter how small dr seuss i teach the smallest students with the biggest enthusiasm \
for learning my students learn in many different ways using all of our senses and multiple intelligences i use a wide range\
of techniques to help all my students succeed students in my class come from a variety of different backgrounds which makes\
for wonderful sharing of experiences and cultures including native americans our school is a caring community of successful \
learners which can be seen through collaborative student project based learning in and out of the classroom kindergarteners \
in my class love to work with hands on materials and have many different opportunities to practice a skill before it is\
mastered having the social skills to work cooperatively with friends is a crucial aspect of the kindergarten curriculum\
montana is the perfect place to learn about agriculture and nutrition my students love to role play in our pretend kitchen\
in the early childhood classroom i have had several kids ask me can we try cooking with real food i will take their idea \
and create common core cooking lessons where we learn important math and writing concepts while cooking delicious healthy \
food for snack time my students will have a grounded appreciation for the work that went into making the food and knowledge \
of where the ingredients came from as well as how it is healthy for their bodies this project would expand our learning of \
nutrition and agricultural cooking recipes by having us peel our own apples to make homemade applesauce make our own bread \
and mix up healthy plants from our classroom garden in the spring we will also create our own cookbooks to be printed and \
shared with families students will gain math and literature skills as well as a life long enjoyment for healthy cooking \
nannan'
ss = sid.polarity_scores(for_sentiment)

for k in ss:
    print('{0}: {1}, '.format(k, ss[k]), end='')

# we can use these 4 things as features/attributes (neg, neu, pos, compound)
# neg: 0.0, neu: 0.753, pos: 0.247, compound: 0.93

D:\installed\Anaconda3\lib\site-packages\nltk\twitter\__init__.py:20: UserWarning:

The twython library has not been installed. Some functionality from the twitter package will not be available.

neg: 0.01, neu: 0.745, pos: 0.245, compound: 0.9975, 

[nltk_data] Downloading package vader_lexicon to C:\Users\KALYAN
[nltk_data]     SRINIVAS\AppData\Roaming\nltk_data...

neg: 0.01, neu: 0.745, pos: 0.245, compound: 0.9975, 

Assignment 5: Logistic Regression

1. [Task-1] Logistic Regression(either SGDClassifier with log loss, or LogisticRegression) on these feature sets
   • Set 1: categorical, numerical features + project_title(BOW) + preprocessed_eassay (`BOW with bi-grams` with `min_df=10` and `max_features=5000`)
   • Set 2: categorical, numerical features + project_title(TFIDF)+ preprocessed_eassay (`TFIDF with bi-grams` with `min_df=10` and `max_features=5000`)
   • Set 3: categorical, numerical features + project_title(AVG W2V)+ preprocessed_eassay (AVG W2V)
   • Set 4: categorical, numerical features + project_title(TFIDF W2V)+ preprocessed_essay (TFIDF W2V)


2. Hyper paramter tuning (find best hyper parameters corresponding the algorithm that you choose)
   • Find the best hyper parameter which will give the maximum AUC value
   • Find the best hyper paramter using k-fold cross validation or simple cross validation data
   • Use gridsearch cv or randomsearch cv or you can also write your own for loops to do this task of hyperparameter tuning


3. Representation of results
   • You need to plot the performance of model both on train data and cross validation data for each hyper parameter, like shown in the figure.
   • Once after you found the best hyper parameter, you need to train your model with it, and find the AUC on test data and plot the ROC curve on both train and test.
   • Along with plotting ROC curve, you need to print the confusion matrix with predicted and original labels of test data points. Please visualize your confusion matrices using seaborn heatmaps.


4. [Task-2] Apply Logistic Regression on the below feature set Set 5 by finding the best hyper parameter as suggested in step 2 and step 3.
5. Consider these set of features Set 5 :
   • school_state : categorical data
   • clean_categories : categorical data
   • clean_subcategories : categorical data
   • project_grade_category :categorical data
   • teacher_prefix : categorical data
   • quantity : numerical data
   • teacher_number_of_previously_posted_projects : numerical data
   • price : numerical data
   • sentiment score's of each of the essay : numerical data
   • number of words in the title : numerical data
   • number of words in the combine essays : numerical data
   And apply the Logistic regression on these features by finding the best hyper paramter as suggested in step 2 and step 3
   


6. Conclusion
   • You need to summarize the results at the end of the notebook, summarize it in the table format. To print out a table please refer to this prettytable library link

Note: Data Leakage

1. There will be an issue of data-leakage if you vectorize the entire data and then split it into train/cv/test.
2. To avoid the issue of data-leakage, make sure to split your data first and then vectorize it.
3. While vectorizing your data, apply the method fit_transform() on you train data, and apply the method transform() on cv/test data.
4. For more details please go through this link.

2. Logistic Regression

2.1 Splitting data into Train and cross validation(or test): Stratified Sampling

# please write all the code with proper documentation, and proper titles for each subsection
# go through documentations and blogs before you start coding
# first figure out what to do, and then think about how to do.
# reading and understanding error messages will be very much helpfull in debugging your code
# when you plot any graph make sure you use 
    # a. Title, that describes your plot, this will be very helpful to the reader
    # b. Legends if needed
    # c. X-axis label
    # d. Y-axis label

project_data.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 109248 entries, 0 to 109247
Data columns (total 20 columns):
 #   Column                                        Non-Null Count   Dtype  
---  ------                                        --------------   -----  
 0   Unnamed: 0                                    109248 non-null  int64  
 1   id                                            109248 non-null  object 
 2   teacher_id                                    109248 non-null  object 
 3   school_state                                  109248 non-null  object 
 4   project_submitted_datetime                    109248 non-null  object 
 5   project_essay_1                               109248 non-null  object 
 6   project_essay_2                               109248 non-null  object 
 7   project_essay_3                               3758 non-null    object 
 8   project_essay_4                               3758 non-null    object 
 9   project_resource_summary                      109248 non-null  object 
 10  teacher_number_of_previously_posted_projects  109248 non-null  int64  
 11  project_is_approved                           109248 non-null  int64  
 12  clean_categories                              109248 non-null  object 
 13  clean_subcategories                           109248 non-null  object 
 14  clean_teacher_prefix                          109248 non-null  object 
 15  clean_project_grade_category                  109248 non-null  object 
 16  price                                         109248 non-null  float64
 17  quantity                                      109248 non-null  int64  
 18  preprocessed_essays                           109248 non-null  object 
 19  preprocessed_titles                           109248 non-null  object 
dtypes: float64(1), int64(4), object(15)
memory usage: 17.5+ MB

we are going to consider

   - school_state : categorical data
   - clean_categories : categorical data
   - clean_subcategories : categorical data
   - project_grade_category : categorical data
   - teacher_prefix : categorical data

   - project_title : text data
   - Essay : text data

   - quantity : numerical 
   - teacher_number_of_previously_posted_projects : numerical
   - price : numerical

data1 = project_data.drop(['Unnamed: 0', 'id','project_submitted_datetime','project_essay_1','project_essay_2','project_essay_3','project_essay_4','project_resource_summary','teacher_id'], axis = 1)

data1.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 109248 entries, 0 to 109247
Data columns (total 11 columns):
 #   Column                                        Non-Null Count   Dtype  
---  ------                                        --------------   -----  
 0   school_state                                  109248 non-null  object 
 1   teacher_number_of_previously_posted_projects  109248 non-null  int64  
 2   project_is_approved                           109248 non-null  int64  
 3   clean_categories                              109248 non-null  object 
 4   clean_subcategories                           109248 non-null  object 
 5   clean_teacher_prefix                          109248 non-null  object 
 6   clean_project_grade_category                  109248 non-null  object 
 7   price                                         109248 non-null  float64
 8   quantity                                      109248 non-null  int64  
 9   preprocessed_essays                           109248 non-null  object 
 10  preprocessed_titles                           109248 non-null  object 
dtypes: float64(1), int64(3), object(7)
memory usage: 10.0+ MB

data1 = data1[:50000]

y = data1['project_is_approved'].values
X = data1.drop(['project_is_approved'], axis=1)
X.head(1)

	school_state	teacher_number_of_previously_posted_projects	clean_categories	clean_subcategories	clean_teacher_prefix	clean_project_grade_category	price	quantity	preprocessed_essays	preprocessed_titles
0	IN	0	Literacy_Language	ESL Literacy	Mrs	PreK-2	154.6	23	my students english learners working english s...	educational support english learners home

# train test split
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, stratify=y)
X_train, X_cv, y_train, y_cv = train_test_split(X_train, y_train, test_size=0.33, stratify=y_train)

2.2 Make Data Model Ready: encoding numerical, categorical features

# please write all the code with proper documentation, and proper titles for each subsection
# go through documentations and blogs before you start coding 
# first figure out what to do, and then think about how to do.
# reading and understanding error messages will be very much helpfull in debugging your code
# make sure you featurize train and test data separatly

# when you plot any graph make sure you use 
    # a. Title, that describes your plot, this will be very helpful to the reader
    # b. Legends if needed
    # c. X-axis label
    # d. Y-axis label

2.2.1 Numerical features

1. teacher_number_of_previously_posted_projects
2. price
3. quantity

2.2.1.1 Teacher number of previously posted projects

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['teacher_number_of_previously_posted_projects'].values.reshape(1,-1))

X_train_TPPP_norm = normalizer.transform(X_train['teacher_number_of_previously_posted_projects'].values.reshape(1,-1))
X_cv_TPPP_norm = normalizer.transform(X_cv['teacher_number_of_previously_posted_projects'].values.reshape(1,-1))
X_test_TPPP_norm = normalizer.transform(X_test['teacher_number_of_previously_posted_projects'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_TPPP_norm.shape, y_train.shape)
print(X_cv_TPPP_norm.shape, y_cv.shape)
print(X_test_TPPP_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of teacher number of previously posted projects")

X_train_TPPP_norm = X_train_TPPP_norm.transpose()
X_cv_TPPP_norm = X_cv_TPPP_norm.transpose()
X_test_TPPP_norm = X_test_TPPP_norm.transpose()

print("After transpose")
print(X_train_TPPP_norm.shape, y_train.shape)
print(X_cv_TPPP_norm.shape, y_cv.shape)
print(X_test_TPPP_norm.shape, y_test.shape)
print("="*100)

Transpose of teacher number of previously posted projects
After transpose
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

2.2.1.2 price

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['price'].values.reshape(1,-1))

X_train_price_norm = normalizer.transform(X_train['price'].values.reshape(1,-1))
X_cv_price_norm = normalizer.transform(X_cv['price'].values.reshape(1,-1))
X_test_price_norm = normalizer.transform(X_test['price'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_price_norm.shape, y_train.shape)
print(X_cv_price_norm.shape, y_cv.shape)
print(X_test_price_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of price")

X_train_price_norm = X_train_price_norm.transpose()
X_cv_price_norm = X_cv_price_norm.transpose()
X_test_price_norm = X_test_price_norm.transpose()


print("After vectorizations")
print(X_train_price_norm.shape, y_train.shape)
print(X_cv_price_norm.shape, y_cv.shape)
print(X_test_price_norm.shape, y_test.shape)
print("="*100)

Transpose of price
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

2.2.1.3 quantity

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['quantity'].values.reshape(1,-1))

X_train_quantity_norm = normalizer.transform(X_train['quantity'].values.reshape(1,-1))
X_cv_quantity_norm = normalizer.transform(X_cv['quantity'].values.reshape(1,-1))
X_test_quantity_norm = normalizer.transform(X_test['quantity'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_quantity_norm.shape, y_train.shape)
print(X_cv_quantity_norm.shape, y_cv.shape)
print(X_test_quantity_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of Quantity")

X_train_quantity_norm = X_train_quantity_norm.transpose()
X_cv_quantity_norm = X_cv_quantity_norm.transpose()
X_test_quantity_norm = X_test_quantity_norm.transpose()


print("After vectorizations")
print(X_train_quantity_norm.shape, y_train.shape)
print(X_cv_quantity_norm.shape, y_cv.shape)
print(X_test_quantity_norm.shape, y_test.shape)
print("="*100)

Transpose of Quantity
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

2.2.2 Categorical Data

Categorical Features for vectorization

1. Clean Categories
2. Clean Sub Categories
3. School State
4. Teacher Prefix
5. Project grade category

2.2.2.1 Clean Categories

vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(X_train['clean_categories'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_CC_ohe = vectorizer.transform(X_train['clean_categories'].values)
X_cv_CC_ohe = vectorizer.transform(X_cv['clean_categories'].values)
X_test_CC_ohe = vectorizer.transform(X_test['clean_categories'].values)

print("After vectorizations")
print(X_train_CC_ohe.shape, y_train.shape)
print(X_cv_CC_ohe.shape, y_cv.shape)
print(X_test_CC_ohe.shape, y_test.shape)
print(vectorizer.get_feature_names())
print("="*100)

After vectorizations
(22445, 9) (22445,)
(11055, 9) (11055,)
(16500, 9) (16500,)
['Warmth', 'Care_Hunger', 'History_Civics', 'Music_Arts', 'AppliedLearning', 'SpecialNeeds', 'Health_Sports', 'Math_Science', 'Literacy_Language']
====================================================================================================

2.2.2.2 Clean Sub Categories

vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(X_train['clean_subcategories'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_CSC_ohe = vectorizer.transform(X_train['clean_subcategories'].values)
X_cv_CSC_ohe = vectorizer.transform(X_cv['clean_subcategories'].values)
X_test_CSC_ohe = vectorizer.transform(X_test['clean_subcategories'].values)

print("After vectorizations")
print(X_train_CSC_ohe.shape, y_train.shape)
print(X_cv_CSC_ohe.shape, y_cv.shape)
print(X_test_CSC_ohe.shape, y_test.shape)
print(vectorizer.get_feature_names())
print("="*100)

After vectorizations
(22445, 30) (22445,)
(11055, 30) (11055,)
(16500, 30) (16500,)
['Economics', 'CommunityService', 'FinancialLiteracy', 'ParentInvolvement', 'Extracurricular', 'Civics_Government', 'ForeignLanguages', 'NutritionEducation', 'Warmth', 'Care_Hunger', 'SocialSciences', 'PerformingArts', 'CharacterEducation', 'TeamSports', 'Other', 'College_CareerPrep', 'Music', 'History_Geography', 'Health_LifeScience', 'EarlyDevelopment', 'ESL', 'Gym_Fitness', 'EnvironmentalScience', 'VisualArts', 'Health_Wellness', 'AppliedSciences', 'SpecialNeeds', 'Literature_Writing', 'Mathematics', 'Literacy']
====================================================================================================

2.2.2.3 School State

vectorizer = CountVectorizer(vocabulary=list(sorted_school_state_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(X_train['school_state'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_state_ohe = vectorizer.transform(X_train['school_state'].values)
X_cv_state_ohe = vectorizer.transform(X_cv['school_state'].values)
X_test_state_ohe = vectorizer.transform(X_test['school_state'].values)

print("After vectorizations")
print(X_train_state_ohe.shape, y_train.shape)
print(X_cv_state_ohe.shape, y_cv.shape)
print(X_test_state_ohe.shape, y_test.shape)
print(vectorizer.get_feature_names())
print("="*100)

After vectorizations
(22445, 51) (22445,)
(11055, 51) (11055,)
(16500, 51) (16500,)
['VT', 'WY', 'ND', 'MT', 'RI', 'SD', 'NE', 'DE', 'AK', 'NH', 'WV', 'ME', 'HI', 'DC', 'NM', 'KS', 'IA', 'ID', 'AR', 'CO', 'MN', 'OR', 'KY', 'MS', 'NV', 'MD', 'CT', 'TN', 'UT', 'AL', 'WI', 'VA', 'AZ', 'NJ', 'OK', 'WA', 'MA', 'LA', 'OH', 'MO', 'IN', 'PA', 'MI', 'SC', 'GA', 'IL', 'NC', 'FL', 'NY', 'TX', 'CA']
====================================================================================================

2.2.2.4 Teacher prefix

vectorizer = CountVectorizer(vocabulary=list(sorted_teacher_prefix_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(X_train['clean_teacher_prefix'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_teacher_ohe = vectorizer.transform(X_train['clean_teacher_prefix'].values)
X_cv_teacher_ohe = vectorizer.transform(X_cv['clean_teacher_prefix'].values)
X_test_teacher_ohe = vectorizer.transform(X_test['clean_teacher_prefix'].values)

print("After vectorizations")
print(X_train_teacher_ohe.shape, y_train.shape)
print(X_cv_teacher_ohe.shape, y_cv.shape)
print(X_test_teacher_ohe.shape, y_test.shape)
print(vectorizer.get_feature_names())
print("="*100)

After vectorizations
(22445, 5) (22445,)
(11055, 5) (11055,)
(16500, 5) (16500,)
['Dr', 'Teacher', 'Mr', 'Ms', 'Mrs']
====================================================================================================

2.2.2.5 Project Grade category

vectorizer = CountVectorizer(vocabulary=list(sorted_project_grade_category_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(X_train['clean_project_grade_category'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_grade_ohe = vectorizer.transform(X_train['clean_project_grade_category'].values)
X_cv_grade_ohe = vectorizer.transform(X_cv['clean_project_grade_category'].values)
X_test_grade_ohe = vectorizer.transform(X_test['clean_project_grade_category'].values)

print("After vectorizations")
print(X_train_grade_ohe.shape, y_train.shape)
print(X_cv_grade_ohe.shape, y_cv.shape)
print(X_test_grade_ohe.shape, y_test.shape)
print(vectorizer.get_feature_names())
print("="*100)

After vectorizations
(22445, 4) (22445,)
(11055, 4) (11055,)
(16500, 4) (16500,)
['9-12', '6-8', '3-5', 'PreK-2']
====================================================================================================

data1['clean_project_grade_category'].unique()

array(['PreK-2', '6-8', '3-5', '9-12'], dtype=object)

2.3 Make Data Model Ready: encoding eassay, and project_title

# please write all the code with proper documentation, and proper titles for each subsection
# go through documentations and blogs before you start coding
# first figure out what to do, and then think about how to do.
# reading and understanding error messages will be very much helpfull in debugging your code
# make sure you featurize train and test data separatly

# when you plot any graph make sure you use 
    # a. Title, that describes your plot, this will be very helpful to the reader
    # b. Legends if needed
    # c. X-axis label
    # d. Y-axis label

Ecoding Essay and Project title

2.3.1 BOW
2.3.2 TFIDF
2.3.3 AVG W2V
2.3.4 TFIDF W2V

2.3.1 BOW Essays and Title

2.3.1.1 BOW Essay

print(X_train.shape, y_train.shape)
print(X_cv.shape, y_cv.shape)
print(X_test.shape, y_test.shape)

print("="*100)


vectorizer = CountVectorizer(min_df=10,ngram_range=(2,2), max_features=5000)
vectorizer.fit(X_train['preprocessed_essays'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_essay_bow = vectorizer.transform(X_train['preprocessed_essays'].values)
X_cv_essay_bow = vectorizer.transform(X_cv['preprocessed_essays'].values)
X_test_essay_bow = vectorizer.transform(X_test['preprocessed_essays'].values)

print("After vectorizations")
print(X_train_essay_bow.shape, y_train.shape)
print(X_cv_essay_bow.shape, y_cv.shape)
print(X_test_essay_bow.shape, y_test.shape)
print("="*100)

(22445, 10) (22445,)
(11055, 10) (11055,)
(16500, 10) (16500,)
====================================================================================================
After vectorizations
(22445, 5000) (22445,)
(11055, 5000) (11055,)
(16500, 5000) (16500,)
====================================================================================================

2.3.1.2 BOW Title

print(X_train.shape, y_train.shape)
print(X_cv.shape, y_cv.shape)
print(X_test.shape, y_test.shape)

print("="*100)


vectorizer = CountVectorizer(min_df=10,ngram_range=(2,2), max_features=5000)
vectorizer.fit(X_train['preprocessed_titles'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_title_bow = vectorizer.transform(X_train['preprocessed_titles'].values)
X_cv_title_bow = vectorizer.transform(X_cv['preprocessed_titles'].values)
X_test_title_bow = vectorizer.transform(X_test['preprocessed_titles'].values)

print("After vectorizations")
print(X_train_title_bow.shape, y_train.shape)
print(X_cv_title_bow.shape, y_cv.shape)
print(X_test_title_bow.shape, y_test.shape)
print("="*100)

(22445, 10) (22445,)
(11055, 10) (11055,)
(16500, 10) (16500,)
====================================================================================================
After vectorizations
(22445, 628) (22445,)
(11055, 628) (11055,)
(16500, 628) (16500,)
====================================================================================================

2.3.2 TF IDF Essay and Title

2.3.2.1 TF IDF Essay

from sklearn.feature_extraction.text import TfidfVectorizer

print(X_train.shape, y_train.shape)
print(X_cv.shape, y_cv.shape)
print(X_test.shape, y_test.shape)

print("="*100)


vectorizer = TfidfVectorizer(min_df=10,ngram_range=(2,2), max_features=5000)
vectorizer.fit(X_train['preprocessed_essays'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_essay_tfidf = vectorizer.transform(X_train['preprocessed_essays'].values)
X_cv_essay_tfidf = vectorizer.transform(X_cv['preprocessed_essays'].values)
X_test_essay_tfidf = vectorizer.transform(X_test['preprocessed_essays'].values)

print("After vectorizations")
print(X_train_essay_tfidf.shape, y_train.shape)
print(X_cv_essay_tfidf.shape, y_cv.shape)
print(X_test_essay_tfidf.shape, y_test.shape)
print("="*100)

(22445, 10) (22445,)
(11055, 10) (11055,)
(16500, 10) (16500,)
====================================================================================================
After vectorizations
(22445, 5000) (22445,)
(11055, 5000) (11055,)
(16500, 5000) (16500,)
====================================================================================================

2.3.2.2 TF IDF Title

print(X_train.shape, y_train.shape)
print(X_cv.shape, y_cv.shape)
print(X_test.shape, y_test.shape)

print("="*100)


vectorizer = TfidfVectorizer(min_df=10,ngram_range=(2,2), max_features=5000)
vectorizer.fit(X_train['preprocessed_titles'].values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector
X_train_title_tfidf = vectorizer.transform(X_train['preprocessed_titles'].values)
X_cv_title_tfidf = vectorizer.transform(X_cv['preprocessed_titles'].values)
X_test_title_tfidf = vectorizer.transform(X_test['preprocessed_titles'].values)

print("After vectorizations")
print(X_train_title_tfidf.shape, y_train.shape)
print(X_cv_title_tfidf.shape, y_cv.shape)
print(X_test_title_tfidf.shape, y_test.shape)
print("="*100)

(22445, 10) (22445,)
(11055, 10) (11055,)
(16500, 10) (16500,)
====================================================================================================
After vectorizations
(22445, 628) (22445,)
(11055, 628) (11055,)
(16500, 628) (16500,)
====================================================================================================

2.3.3 AVG W2V Essay and Title

2.3.3.1 AVG W2V Essay

# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
# make sure you have the glove_vectors file
with open('../glove_vectors', 'rb') as f:
    model = pickle.load(f)
    glove_words =  set(model.keys())

# average Word2Vec
# compute average word2vec for each review.
avg_w2v_vectors_train = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_train['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_train.append(vector)

print(len(avg_w2v_vectors_train))
print(len(avg_w2v_vectors_train[0]))
print(avg_w2v_vectors_train[0])

100%|██████████| 22445/22445 [00:09<00:00, 2418.28it/s]

22445
300
[-1.81056382e-02  1.85787859e-02  8.02200592e-03 -8.32963112e-02
  3.81789230e-02 -5.63705612e-02 -3.02692118e+00  2.32871658e-02
 -1.42638191e-02 -7.91525276e-02  5.29135500e-02 -4.73494486e-02
  1.10784000e-01 -1.39750720e-01 -4.06088651e-02 -3.49337092e-02
 -1.62968197e-02 -5.07861593e-02  9.10414599e-02  2.41862605e-02
  2.98720987e-02 -4.40122368e-04 -3.77595052e-02 -2.28924336e-02
 -1.81628539e-02 -6.43741553e-02  1.48104853e-01 -5.48663816e-02
 -5.84205178e-02 -1.33388737e-01 -2.41679681e-01 -8.77509066e-02
  7.39125158e-02  7.38188217e-02 -6.07830368e-02 -6.69004599e-02
 -2.92136145e-02 -2.20923651e-02 -1.67388013e-02 -1.17562066e-02
 -8.65316395e-02  6.42400868e-02 -4.86157623e-02 -1.16614553e-01
  4.39485586e-02 -5.98070000e-02  8.60767013e-02 -3.82555868e-02
 -6.36866250e-03 -1.55155081e-01  8.65253355e-03 -3.91407408e-02
  2.39282546e-02 -2.36819145e-03  3.43587428e-02 -7.54512962e-02
  4.00895888e-02 -5.66493421e-05 -1.13641489e-01  9.22826625e-02
 -4.39659086e-02 -7.46109845e-02  1.12309228e-01 -4.93016513e-02
 -1.26157559e-01  1.11218863e-01  6.01222519e-02 -5.03954776e-02
  1.14845042e-01 -1.14323824e-01 -1.42451474e-01  2.01336776e-03
 -1.13392283e-02 -8.67249750e-02 -3.98609046e-02 -2.07408928e-01
  1.30471268e-02  3.70433717e-02  9.34196855e-02 -7.02459288e-02
  3.54300820e-02 -2.56097872e-01 -3.42834099e-02 -4.63499934e-02
 -1.06201800e-01  8.06505079e-02  1.43964434e-01 -4.77907198e-02
  1.38064498e-01 -3.11918809e-02  3.15169912e-02 -4.92050263e-03
  7.81461316e-03  4.72643539e-02 -6.75543030e-02 -1.80513704e-01
 -2.25664961e+00  4.84546301e-02  1.05425723e-01  1.16757968e-01
 -1.17735961e-01 -8.58626928e-02  1.20768795e-01 -3.88365263e-02
  1.17130429e-01  1.41773257e-02  3.02586572e-02 -1.27251211e-01
 -3.62788571e-02  1.01552051e-01 -5.46537303e-03  2.34891412e-02
  3.40654140e-02  2.67138457e-01  1.91061316e-03  6.19906118e-02
 -2.19711411e-01  1.27568586e-02  9.97981739e-02  1.90838205e-03
  5.08465132e-04  7.26526355e-02  7.38808920e-02 -6.15298724e-02
  9.57686520e-02  1.63401451e-02  6.37212451e-02 -2.05026868e-02
  2.60540592e-02  1.72860780e-01  3.76683176e-02  3.07486645e-02
 -6.37487507e-02 -6.06424099e-02  2.55144664e-02 -1.20271561e-01
  1.46133883e-01 -6.26942770e-02  1.10942816e-01  3.09165661e-01
  6.89717829e-02  2.25713224e-03  3.08901612e-02 -1.51810382e-02
 -6.04713758e-02 -1.83180789e-02  6.45223592e-02 -4.87112908e-02
  2.23118474e-01 -1.45735435e-02 -4.13769441e-02 -4.82520039e-02
 -2.59495414e-02 -5.53969850e-02  2.85938536e-02  9.15658816e-03
  8.63843158e-03 -4.16889474e-03 -5.31070913e-02  3.97150270e-02
  3.33078546e-02  2.75351171e-02 -1.34087539e-02 -2.80827112e-02
 -6.81389164e-02  5.10345329e-02 -9.06004118e-02  3.85811237e-02
  5.76649971e-02 -2.38291220e-02  1.80143164e-02 -9.03551579e-03
 -5.37487908e-02 -1.25199855e-01 -3.54340197e-02  3.06774289e-02
 -2.60475770e-02  2.05901382e-03 -1.81464716e-01 -2.67299580e-02
  2.38467592e-02  2.60481217e-01 -4.53077362e-02  1.39267162e-02
 -5.00981875e-02 -3.88024496e-02 -7.69773847e-02 -4.66132048e-02
  1.12805682e-01  3.40712120e-02 -7.15964474e-03 -9.35880493e-02
 -1.09866130e-01 -3.76330851e-02 -1.61644934e-02 -1.55948020e-02
  3.54425066e-03  4.17711493e-02  7.64680638e-02  5.50856274e-02
  1.48950456e-01 -5.81642724e-02 -9.60174125e-03  1.02514975e-01
 -9.28153411e-02  1.43380822e-02  1.10363392e-01 -1.54230274e-01
  1.52852347e-01  6.14320901e-02 -1.79965526e-03  4.34496026e-02
 -3.61592528e-02 -1.54023204e-01 -1.09210647e-01 -1.80599605e-02
 -8.96003303e-02 -6.20648572e-02 -4.50942967e-02 -2.24179731e-03
 -1.15454105e-01 -5.59059901e-02 -1.13956885e-01 -1.76420053e-01
 -2.15369061e+00  1.05254500e-01 -5.44177956e-02  4.57045264e-02
 -7.57690309e-02 -7.87857322e-02  8.27827961e-03 -4.13595007e-02
 -3.66299138e-02 -6.27829934e-02 -9.95262082e-02  2.33578553e-03
  1.06652017e-01 -6.32505350e-02  2.38090204e-02  9.82542474e-02
 -3.77676276e-02  2.32552954e-02 -2.11784053e-01  3.55360691e-02
 -1.97750237e-02  1.92008322e-02 -2.47952658e-02 -1.01955404e-01
  3.74933895e-02 -2.34210141e-02  1.69430691e-02  1.37341568e-01
  9.23217776e-02 -5.83219618e-02  1.16385039e-01 -2.25385395e-02
  8.30138499e-02 -1.23081441e-02  1.24054683e-01 -1.16845710e-01
  1.28277717e-02  2.05918257e-02  3.14728526e-02 -1.76624533e-02
  9.54082711e-02 -1.57384462e-01 -9.34587087e-02  1.70310079e-02
  4.94733783e-02  1.09127035e-01 -5.45181737e-02 -1.92485013e-02
 -1.88040450e-01  7.78147625e-02 -7.93638493e-02  8.29783336e-02
  9.38740237e-02 -5.61750566e-03 -6.16241375e-02  9.13889553e-02
  1.65522000e-01 -6.11511770e-02 -2.83500276e-02  1.04832883e-01
 -6.23275395e-03  1.45701301e-01  7.13180461e-03 -1.40408711e-02
  6.11556447e-02 -1.62285849e-02  2.00333211e-02 -4.69465724e-02
 -4.57653445e-02 -8.10039661e-02 -1.30848618e-02  2.17784201e-02
  1.56738092e-02  1.87370527e-01  1.32548711e-01  4.63472020e-02]

avg_w2v_vectors_cv = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_cv['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_cv.append(vector)

100%|██████████| 11055/11055 [00:04<00:00, 2428.41it/s]

avg_w2v_vectors_test = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_test['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_test.append(vector)

100%|██████████| 16500/16500 [00:07<00:00, 2278.52it/s]

2.3.3.2 AVG W2V Title

# average Word2Vec
# compute average word2vec for each review.
avg_w2v_vectors_train_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_train['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_train_title.append(vector)

print(len(avg_w2v_vectors_train_title))
print(len(avg_w2v_vectors_train_title[0]))
print(avg_w2v_vectors_train_title[0])

100%|██████████| 22445/22445 [00:00<00:00, 45963.70it/s]

22445
300
[-7.46400000e-02  1.22166667e-01 -1.77160000e-01 -9.14593333e-02
  7.43523333e-02 -3.64743333e-01 -3.24416667e+00 -1.24890000e-01
  3.20666667e-03 -2.47513333e-01  3.96360000e-01 -1.37836667e-01
 -4.38466667e-02 -3.77453333e-01 -2.26230000e-01 -1.40898000e-01
 -3.14900000e-02  1.95137667e-02 -6.72833333e-02  2.84846667e-01
 -1.53411000e-01 -9.58300000e-02  9.19963333e-02 -4.36800000e-02
  4.71666667e-04 -1.55254300e-01  4.16944000e-01 -3.05133333e-02
  1.17966667e-02 -1.23728667e-01 -4.68513333e-01 -6.75266667e-02
  2.77133333e-02  1.91263667e-01 -2.62876000e-01 -6.36866667e-02
  5.14786667e-02 -2.48933333e-02 -1.39532667e-01 -3.65356667e-02
 -2.98676667e-01  2.13733333e-01  1.67173000e-01 -2.25343433e-01
  3.40833333e-02 -1.38386667e-01  3.21806667e-01 -4.53590000e-02
  7.11123333e-02 -1.83020000e-01 -1.09260000e-01  1.12607667e-01
  1.59501667e-01  1.18681333e-01 -5.71000000e-02 -1.72638000e-01
 -2.06206667e-01 -4.25000000e-02  1.19436667e-01  4.04083333e-01
 -2.25556667e-01 -6.56635667e-03  1.25213333e-01 -5.58166667e-02
 -9.98900000e-02 -1.32353333e-01 -1.73553333e-01 -1.37396333e-01
  1.13220000e-01 -2.17070000e-02 -1.16148333e-01  9.46010000e-02
  1.05372333e-01 -7.80823333e-02  1.06652333e-01 -3.68313333e-01
  1.06980000e-01 -1.34726667e-02  1.28333333e-02 -1.66035000e-01
  1.25800667e-01 -4.00580000e-01 -2.31803333e-01  8.11216667e-02
 -1.97846667e-01  1.35606667e-01  3.09318000e-01 -1.56812000e-01
  2.27065333e-01 -4.30296667e-02  1.41181333e-01  3.98926667e-02
  1.04339667e-01  2.39186000e-01 -3.80113333e-01 -1.15072333e-01
 -2.31620000e+00 -1.06233333e-02  1.76133333e-03  3.91266667e-02
 -2.47966667e-02 -2.73833333e-01  1.13833333e-02 -9.04046667e-02
  1.06583333e-01  9.05523333e-02  4.45923333e-02 -3.30250000e-01
  1.01566667e-01  1.67980000e-01 -1.31998667e-01  1.43223333e-01
 -8.35756667e-02  4.66623333e-01  1.01903333e-01  5.65316667e-01
 -1.62340667e-01  1.36238233e-01  5.04170000e-01  7.64200000e-02
 -2.48166667e-02  4.07666667e-01  2.08154333e-01 -1.57109333e-01
  3.19366667e-01  2.09970600e-01  3.30186667e-01  5.09990000e-02
 -2.50833333e-02  2.53933333e-01  4.65066667e-02  1.09100000e-01
 -7.50866667e-03 -8.92933333e-02 -5.66433333e-03 -9.73396667e-02
  4.99500000e-02 -1.38866667e-02  1.60373333e-01  8.09070333e-01
  2.18844333e-01 -2.12059667e-01 -1.92143333e-01  1.99526667e-01
 -1.03400333e-01 -8.79403333e-02  1.10116667e-02 -7.04163333e-02
 -5.96697000e-02  8.74070000e-02  1.11333333e-02  9.90566667e-02
  3.29103333e-01 -8.68122667e-02  4.53000000e-03 -3.46522000e-01
 -5.28228000e-02 -9.87400000e-02  5.34233333e-02  5.16933333e-02
 -1.18356667e-01  1.81843000e-01 -1.09173333e-01 -3.96083333e-02
  1.01094000e-01  2.27110000e-01 -6.44053333e-02  1.80280000e-01
  1.81425000e-01 -2.97646667e-01  2.37575667e-01  1.52993667e-01
 -4.05740000e-02 -4.27670000e-01 -1.41356333e-01  1.86466667e-02
 -4.73033333e-02 -2.61016667e-01 -3.85243333e-01 -1.90686000e-01
 -7.11493333e-02  5.35660000e-01 -1.49166667e-02  3.25420000e-02
  1.17566667e-02  1.68869000e-01 -3.23676667e-01  1.81256667e-02
  1.77487000e-01 -1.17078333e-01  5.50243333e-02 -1.20486667e-01
 -3.73766667e-01 -1.84884333e-01 -1.66073333e-01 -3.33718333e-02
 -3.92496667e-01  1.87206667e-01 -9.52540000e-02  1.19205333e-01
  2.79591333e-01 -2.24760000e-01 -3.03266667e-02  2.41365667e-01
 -3.18494333e-01 -1.81900000e-02  7.91590000e-02 -2.41422067e-01
  5.28286667e-01 -8.98177000e-02  1.06021333e-01  1.61886667e-01
 -9.51930000e-02 -1.01511333e-01 -4.70599023e-02  8.38573333e-02
 -2.68823333e-01  7.90800000e-02 -1.41200000e-01 -1.86933000e-01
  5.94266667e-03  1.97166667e-01 -1.31300333e-01 -1.52923333e-01
 -2.94996667e+00 -1.82753333e-01 -1.39843663e-01 -6.50220000e-02
  1.18590000e-01 -1.70672667e-01 -3.21566667e-02  1.15526667e-01
 -3.50023333e-02 -2.23545000e-01 -1.42237333e-01 -3.81326667e-02
  2.10989000e-01 -1.91607667e-01  3.79560000e-01  3.08324000e-01
 -2.34946667e-01 -8.36406667e-02 -3.15210000e-01  5.34070000e-01
 -2.31316667e-01  3.63493333e-01  1.71458667e-01 -1.31643333e-01
 -1.84276000e-01  5.67463333e-02 -4.60800000e-02  8.73006667e-02
  2.44176667e-01  1.72602200e-01  2.17212333e-01 -9.08100000e-03
  3.37860000e-01  3.85746667e-02  3.98933333e-02 -4.34016667e-02
 -1.76755000e-01  2.86233333e-01  1.16408667e-01 -1.57438000e-01
  1.86796667e-01  5.53263667e-02 -3.46066667e-01  8.79338667e-02
  4.30023333e-02  3.27428333e-01  7.63733333e-02 -2.94533333e-02
 -3.94330000e-01  2.11124333e-01 -2.39633333e-02 -1.60830000e-02
 -5.91766667e-02  1.20007667e-01 -1.07386667e-01  2.64350000e-01
  4.26680000e-01 -4.04830000e-02 -2.63670000e-01  9.54666667e-02
  1.42130000e-01  1.02530667e-01  5.26060000e-02  4.09990000e-01
  4.70000000e-02 -1.26870000e-01  2.55113333e-01  9.69466667e-02
 -1.70628333e-01 -5.35663333e-02 -7.60686667e-02  3.48946667e-01
  1.67374500e-01  3.72966667e-03  2.51486667e-01  1.31202667e-01]

avg_w2v_vectors_cv_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_cv['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_cv_title.append(vector)

100%|██████████| 11055/11055 [00:00<00:00, 47985.68it/s]

avg_w2v_vectors_test_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_test['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    cnt_words =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if word in glove_words:
            vector += model[word]
            cnt_words += 1
    if cnt_words != 0:
        vector /= cnt_words
    avg_w2v_vectors_test_title.append(vector)

100%|██████████| 16500/16500 [00:00<00:00, 50975.87it/s]

2.3.4 TF IDF W2V Essay and Title

2.3.4.1 TF IDF W2V Essay

# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(X_train['preprocessed_essays'].values)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())

# average Word2Vec
# compute average word2vec for each review.
tfidf_w2v_vectors_train = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_train['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_train.append(vector)

print(len(tfidf_w2v_vectors_train))
print(len(tfidf_w2v_vectors_train[0]))

100%|██████████| 22445/22445 [01:05<00:00, 342.86it/s]

22445
300

tfidf_w2v_vectors_cv = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_cv['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_cv.append(vector)

100%|██████████| 11055/11055 [00:42<00:00, 262.59it/s]

tfidf_w2v_vectors_test = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_test['preprocessed_essays'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_test.append(vector)

100%|██████████| 16500/16500 [01:07<00:00, 243.65it/s]

2.3.4.2 TF IDF W2V Title

# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(X_train['preprocessed_titles'].values)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())

# average Word2Vec
# compute average word2vec for each review.
tfidf_w2v_vectors_train_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_train['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_train_title.append(vector)

print(len(tfidf_w2v_vectors_train_title))
print(len(tfidf_w2v_vectors_train_title[0]))

100%|██████████| 22445/22445 [00:01<00:00, 14705.60it/s]

22445
300

tfidf_w2v_vectors_cv_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_cv['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_cv_title.append(vector)

100%|██████████| 11055/11055 [00:00<00:00, 16735.77it/s]

tfidf_w2v_vectors_test_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(X_test['preprocessed_titles'].values): # for each review/sentence
    vector = np.zeros(300) # as word vectors are of zero length
    tf_idf_weight =0; # num of words with a valid vector in the sentence/review
    for word in sentence.split(): # for each word in a review/sentence
        if (word in glove_words) and (word in tfidf_words):
            vec = model[word] # getting the vector for each word
            # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
            tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
            vector += (vec * tf_idf) # calculating tfidf weighted w2v
            tf_idf_weight += tf_idf
    if tf_idf_weight != 0:
        vector /= tf_idf_weight
    tfidf_w2v_vectors_test_title.append(vector)

100%|██████████| 16500/16500 [00:00<00:00, 17942.96it/s]

Concatinating all the features

1. SET 1 BOW

# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
X_tr_BOW = hstack((X_train_essay_bow, X_train_title_bow, X_train_state_ohe, X_train_teacher_ohe, X_train_grade_ohe, X_train_CSC_ohe, X_train_CC_ohe, X_train_price_norm, X_train_quantity_norm, X_train_TPPP_norm)).tocsr()
X_cr_BOW = hstack((X_cv_essay_bow, X_cv_title_bow, X_cv_state_ohe, X_cv_teacher_ohe, X_cv_grade_ohe, X_cv_CSC_ohe, X_cv_CC_ohe, X_cv_price_norm, X_cv_quantity_norm, X_cv_TPPP_norm)).tocsr()
X_te_BOW = hstack((X_test_essay_bow, X_test_title_bow, X_test_state_ohe, X_test_teacher_ohe, X_test_grade_ohe, X_test_CSC_ohe, X_test_CC_ohe, X_test_price_norm, X_test_quantity_norm, X_test_TPPP_norm)).tocsr()

print("Final Data matrix")
print(X_tr_BOW.shape, y_train.shape)
print(X_cr_BOW.shape, y_cv.shape)
print(X_te_BOW.shape, y_test.shape)
print("="*100)

Final Data matrix
(22445, 5730) (22445,)
(11055, 5730) (11055,)
(16500, 5730) (16500,)
====================================================================================================

2. SET 2 TF IDF

# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
X_tr_TFIDF = hstack((X_train_essay_tfidf, X_train_title_tfidf, X_train_state_ohe, X_train_teacher_ohe, X_train_grade_ohe, X_train_CSC_ohe, X_train_CC_ohe, X_train_price_norm, X_train_quantity_norm, X_train_TPPP_norm)).tocsr()
X_cr_TFIDF = hstack((X_cv_essay_tfidf, X_cv_title_tfidf, X_cv_state_ohe, X_cv_teacher_ohe, X_cv_grade_ohe, X_cv_CSC_ohe, X_cv_CC_ohe, X_cv_price_norm, X_cv_quantity_norm, X_cv_TPPP_norm)).tocsr()
X_te_TFIDF = hstack((X_test_essay_tfidf, X_test_title_tfidf, X_test_state_ohe, X_test_teacher_ohe, X_test_grade_ohe, X_test_CSC_ohe, X_test_CC_ohe, X_test_price_norm, X_test_quantity_norm, X_test_TPPP_norm)).tocsr()

print("Final Data matrix")
print(X_tr_TFIDF.shape, y_train.shape)
print(X_cr_TFIDF.shape, y_cv.shape)
print(X_te_TFIDF.shape, y_test.shape)
print("="*100)

Final Data matrix
(22445, 5730) (22445,)
(11055, 5730) (11055,)
(16500, 5730) (16500,)
====================================================================================================

3. SET 3 AVG W2V

# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
X_tr_AVG_W2V = hstack((avg_w2v_vectors_train, avg_w2v_vectors_train_title, X_train_state_ohe, X_train_teacher_ohe, X_train_grade_ohe, X_train_CSC_ohe, X_train_CC_ohe, X_train_price_norm, X_train_quantity_norm, X_train_TPPP_norm)).tocsr()
X_cr_AVG_W2V = hstack((avg_w2v_vectors_cv, avg_w2v_vectors_cv_title, X_cv_state_ohe, X_cv_teacher_ohe, X_cv_grade_ohe, X_cv_CSC_ohe, X_cv_CC_ohe, X_cv_price_norm, X_cv_quantity_norm, X_cv_TPPP_norm)).tocsr()
X_te_AVG_W2V = hstack((avg_w2v_vectors_test, avg_w2v_vectors_test_title, X_test_state_ohe, X_test_teacher_ohe, X_test_grade_ohe, X_test_CSC_ohe, X_test_CC_ohe, X_test_price_norm, X_test_quantity_norm, X_test_TPPP_norm)).tocsr()

print("Final Data matrix")
print(X_tr_AVG_W2V.shape, y_train.shape)
print(X_cr_AVG_W2V.shape, y_cv.shape)
print(X_te_AVG_W2V.shape, y_test.shape)
print("="*100)

Final Data matrix
(22445, 702) (22445,)
(11055, 702) (11055,)
(16500, 702) (16500,)
====================================================================================================

4. SET 4 TF IDF W2V

# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
X_tr_TFIDF_W2V = hstack((tfidf_w2v_vectors_train, tfidf_w2v_vectors_train_title, X_train_state_ohe, X_train_teacher_ohe, X_train_grade_ohe, X_train_CSC_ohe, X_train_CC_ohe, X_train_price_norm, X_train_quantity_norm, X_train_TPPP_norm)).tocsr()
X_cr_TFIDF_W2V = hstack((tfidf_w2v_vectors_cv, tfidf_w2v_vectors_cv_title, X_cv_state_ohe, X_cv_teacher_ohe, X_cv_grade_ohe, X_cv_CSC_ohe, X_cv_CC_ohe, X_cv_price_norm, X_cv_quantity_norm, X_cv_TPPP_norm)).tocsr()
X_te_TFIDF_W2V = hstack((tfidf_w2v_vectors_test, tfidf_w2v_vectors_test_title, X_test_state_ohe, X_test_teacher_ohe, X_test_grade_ohe, X_test_CSC_ohe, X_test_CC_ohe, X_test_price_norm, X_test_quantity_norm, X_test_TPPP_norm)).tocsr()

print("Final Data matrix")
print(X_tr_TFIDF_W2V.shape, y_train.shape)
print(X_cr_TFIDF_W2V.shape, y_cv.shape)
print(X_te_TFIDF_W2V.shape, y_test.shape)
print("="*100)

Final Data matrix
(22445, 702) (22445,)
(11055, 702) (11055,)
(16500, 702) (16500,)
====================================================================================================

2.4 Appling Logistic Regression on different kind of featurization as mentioned in the instructions

Apply Logistic Regression on different kind of featurization as mentioned in the instructions
For Every model that you work on make sure you do the step 2 and step 3 of instrucations

# please write all the code with proper documentation, and proper titles for each subsection
# go through documentations and blogs before you start coding
# first figure out what to do, and then think about how to do.
# reading and understanding error messages will be very much helpfull in debugging your code
# when you plot any graph make sure you use 
    # a. Title, that describes your plot, this will be very helpful to the reader
    # b. Legends if needed
    # c. X-axis label
    # d. Y-axis label

2.4.1 Applying logistic regression on BOW, SET 1

import warnings
warnings.filterwarnings('ignore')
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
#from sklearn.grid_search import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import learning_curve, GridSearchCV

"""
y_true : array, shape = [n_samples] or [n_samples, n_classes]
True binary labels or binary label indicators.

y_score : array, shape = [n_samples] or [n_samples, n_classes]
Target scores, can either be probability estimates of the positive class, confidence values, or non-thresholded measure of
decisions (as returned by “decision_function” on some classifiers). 
For binary y_true, y_score is supposed to be the score of the class with greater label.

"""



clf = LogisticRegression(class_weight='balanced');
parameters ={'C':[10**-4, 10**-3,10**-2,1,10,100,1000,500,1000,10000]}
sd=GridSearchCV(clf, parameters, cv=5, scoring='roc_auc',return_train_score=True)
sd.fit(X_tr_BOW, y_train);


train_auc= sd.cv_results_['mean_train_score']
train_auc_std= sd.cv_results_['std_train_score']
cv_auc = sd.cv_results_['mean_test_score'] 
cv_auc_std= sd.cv_results_['std_test_score']

plt.plot(parameters['C'], train_auc, label='Train AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],train_auc - train_auc_std,train_auc + train_auc_std,alpha=0.2,color='darkblue')

plt.plot(parameters['C'], cv_auc, label='CV AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],cv_auc - cv_auc_std,cv_auc + cv_auc_std,alpha=0.2,color='darkorange')

plt.scatter(parameters['C'], train_auc, label='Train AUC points')
plt.scatter(parameters['C'], cv_auc, label='CV AUC points')
plt.xscale('log') 

plt.legend()
plt.xlabel("C(1/lambda): hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

##Fitting Model to Hyper-Parameter Curve
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc


neigh = LogisticRegression(C=10**-2,class_weight='balanced');
neigh.fit(X_tr_BOW ,y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

train_fpr, train_tpr, thresholds = roc_curve(y_train, neigh.predict_proba(X_tr_BOW)[:,1])
test_fpr, test_tpr, thresholds = roc_curve(y_test, neigh.predict_proba(X_te_BOW)[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.ylabel("True Positive Rate(TPR)")
plt.xlabel("False Positive Rate(FPR)")
plt.title("ROC PLOTS")
plt.show()

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_train, neigh.predict(X_tr_BOW )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_test, neigh.predict(X_te_BOW )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

2.4.2 Applying Logistic regression on TFIDF, SET 2

import warnings
warnings.filterwarnings('ignore')
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
#from sklearn.grid_search import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import learning_curve, GridSearchCV

"""
y_true : array, shape = [n_samples] or [n_samples, n_classes]
True binary labels or binary label indicators.

y_score : array, shape = [n_samples] or [n_samples, n_classes]
Target scores, can either be probability estimates of the positive class, confidence values, or non-thresholded measure of
decisions (as returned by “decision_function” on some classifiers). 
For binary y_true, y_score is supposed to be the score of the class with greater label.

"""



clf = LogisticRegression(class_weight='balanced');
parameters ={'C':[10**-4, 10**-3,10**-2,1,10,100,1000,500,1000,10000]}
sd=GridSearchCV(clf, parameters, cv=5, scoring='roc_auc',return_train_score=True)
sd.fit(X_tr_TFIDF, y_train);


train_auc= sd.cv_results_['mean_train_score']
train_auc_std= sd.cv_results_['std_train_score']
cv_auc = sd.cv_results_['mean_test_score'] 
cv_auc_std= sd.cv_results_['std_test_score']

plt.plot(parameters['C'], train_auc, label='Train AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],train_auc - train_auc_std,train_auc + train_auc_std,alpha=0.2,color='darkblue')

plt.plot(parameters['C'], cv_auc, label='CV AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],cv_auc - cv_auc_std,cv_auc + cv_auc_std,alpha=0.2,color='darkorange')

plt.scatter(parameters['C'], train_auc, label='Train AUC points')
plt.scatter(parameters['C'], cv_auc, label='CV AUC points')
plt.xscale('log') 

plt.legend()
plt.xlabel("C(1/lambda): hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

##Fitting Model to Hyper-Parameter Curve
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc


neigh = LogisticRegression(C=10**0,class_weight='balanced');
neigh.fit(X_tr_TFIDF ,y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

train_fpr, train_tpr, thresholds = roc_curve(y_train, neigh.predict_proba(X_tr_TFIDF)[:,1])
test_fpr, test_tpr, thresholds = roc_curve(y_test, neigh.predict_proba(X_te_TFIDF)[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.ylabel("True Positive Rate(TPR)")
plt.xlabel("False Positive Rate(FPR)")
plt.title("ROC PLOTS")
plt.show()

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_train, neigh.predict(X_tr_TFIDF )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_test, neigh.predict(X_te_TFIDF )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

2.4.3 Applying Logistic regression on AVG W2V, SET 3

import warnings
warnings.filterwarnings('ignore')
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
#from sklearn.grid_search import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import learning_curve, GridSearchCV

"""
y_true : array, shape = [n_samples] or [n_samples, n_classes]
True binary labels or binary label indicators.

y_score : array, shape = [n_samples] or [n_samples, n_classes]
Target scores, can either be probability estimates of the positive class, confidence values, or non-thresholded measure of
decisions (as returned by “decision_function” on some classifiers). 
For binary y_true, y_score is supposed to be the score of the class with greater label.

"""



clf = LogisticRegression(class_weight='balanced');
parameters ={'C':[10**-4, 10**-3,10**-2,1,10,100,1000,500,1000,10000]}
sd=GridSearchCV(clf, parameters, cv=5, scoring='roc_auc',return_train_score=True)
sd.fit(X_tr_AVG_W2V, y_train);


train_auc= sd.cv_results_['mean_train_score']
train_auc_std= sd.cv_results_['std_train_score']
cv_auc = sd.cv_results_['mean_test_score'] 
cv_auc_std= sd.cv_results_['std_test_score']

plt.plot(parameters['C'], train_auc, label='Train AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],train_auc - train_auc_std,train_auc + train_auc_std,alpha=0.2,color='darkblue')

plt.plot(parameters['C'], cv_auc, label='CV AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],cv_auc - cv_auc_std,cv_auc + cv_auc_std,alpha=0.2,color='darkorange')

plt.scatter(parameters['C'], train_auc, label='Train AUC points')
plt.scatter(parameters['C'], cv_auc, label='CV AUC points')
plt.xscale('log') 

plt.legend()
plt.xlabel("C(1/lambda): hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

##Fitting Model to Hyper-Parameter Curve
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc


neigh = LogisticRegression(C=10**0,class_weight='balanced');
neigh.fit(X_tr_AVG_W2V ,y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

train_fpr, train_tpr, thresholds = roc_curve(y_train, neigh.predict_proba(X_tr_AVG_W2V)[:,1])
test_fpr, test_tpr, thresholds = roc_curve(y_test, neigh.predict_proba(X_te_AVG_W2V)[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.ylabel("True Positive Rate(TPR)")
plt.xlabel("False Positive Rate(FPR)")
plt.title("ROC PLOTS")
plt.show()

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_train, neigh.predict(X_tr_AVG_W2V )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_test, neigh.predict(X_te_AVG_W2V )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

2.4.4 Applying Logistic regression on TFIDF W2v, SET 4

import warnings
warnings.filterwarnings('ignore')
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
#from sklearn.grid_search import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import learning_curve, GridSearchCV

"""
y_true : array, shape = [n_samples] or [n_samples, n_classes]
True binary labels or binary label indicators.

y_score : array, shape = [n_samples] or [n_samples, n_classes]
Target scores, can either be probability estimates of the positive class, confidence values, or non-thresholded measure of
decisions (as returned by “decision_function” on some classifiers). 
For binary y_true, y_score is supposed to be the score of the class with greater label.

"""



clf = LogisticRegression(class_weight='balanced');
parameters ={'C':[10**-4, 10**-3,10**-2,1,10,100,1000,500,1000,10000]}
sd=GridSearchCV(clf, parameters, cv=5, scoring='roc_auc',return_train_score=True)
sd.fit(X_tr_TFIDF_W2V, y_train);


train_auc= sd.cv_results_['mean_train_score']
train_auc_std= sd.cv_results_['std_train_score']
cv_auc = sd.cv_results_['mean_test_score'] 
cv_auc_std= sd.cv_results_['std_test_score']

plt.plot(parameters['C'], train_auc, label='Train AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],train_auc - train_auc_std,train_auc + train_auc_std,alpha=0.2,color='darkblue')

plt.plot(parameters['C'], cv_auc, label='CV AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],cv_auc - cv_auc_std,cv_auc + cv_auc_std,alpha=0.2,color='darkorange')

plt.scatter(parameters['C'], train_auc, label='Train AUC points')
plt.scatter(parameters['C'], cv_auc, label='CV AUC points')
plt.xscale('log') 

plt.legend()
plt.xlabel("C(1/lambda): hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

##Fitting Model to Hyper-Parameter Curve
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc


neigh = LogisticRegression(C=10**-2,class_weight='balanced');
neigh.fit(X_tr_TFIDF_W2V ,y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

train_fpr, train_tpr, thresholds = roc_curve(y_train, neigh.predict_proba(X_tr_TFIDF_W2V)[:,1])
test_fpr, test_tpr, thresholds = roc_curve(y_test, neigh.predict_proba(X_te_TFIDF_W2V)[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.ylabel("True Positive Rate(TPR)")
plt.xlabel("False Positive Rate(FPR)")
plt.title("ROC PLOTS")
plt.show()

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_train, neigh.predict(X_tr_TFIDF_W2V )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_test, neigh.predict(X_te_TFIDF_W2V )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

2.5 Logistic Regression with added Features `Set 5`

# please write all the code with proper documentation, and proper titles for each subsection
# go through documentations and blogs before you start coding
# first figure out what to do, and then think about how to do.
# reading and understanding error messages will be very much helpfull in debugging your code
# when you plot any graph make sure you use 
    # a. Title, that describes your plot, this will be very helpful to the reader
    # b. Legends if needed
    # c. X-axis label
    # d. Y-axis label

data1.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 50000 entries, 0 to 49999
Data columns (total 11 columns):
 #   Column                                        Non-Null Count  Dtype  
---  ------                                        --------------  -----  
 0   school_state                                  50000 non-null  object 
 1   teacher_number_of_previously_posted_projects  50000 non-null  int64  
 2   project_is_approved                           50000 non-null  int64  
 3   clean_categories                              50000 non-null  object 
 4   clean_subcategories                           50000 non-null  object 
 5   clean_teacher_prefix                          50000 non-null  object 
 6   clean_project_grade_category                  50000 non-null  object 
 7   price                                         50000 non-null  float64
 8   quantity                                      50000 non-null  int64  
 9   preprocessed_essays                           50000 non-null  object 
 10  preprocessed_titles                           50000 non-null  object 
dtypes: float64(1), int64(3), object(7)
memory usage: 4.6+ MB

import nltk
from nltk.sentiment.vader import SentimentIntensityAnalyzer

# import nltk
nltk.download('vader_lexicon')

sid = SentimentIntensityAnalyzer()

neg = []
pos = []
neu = []
compound = []


for a in tqdm(data1["preprocessed_essays"]) :
    b = sid.polarity_scores(a)['neg']
    c = sid.polarity_scores(a)['pos']
    d = sid.polarity_scores(a)['neu']
    e = sid.polarity_scores(a)['compound']
    neg.append(b)
    pos.append(c)
    neu.append(d)
    compound.append(e)

[nltk_data] Downloading package vader_lexicon to C:\Users\KALYAN
[nltk_data]     SRINIVAS\AppData\Roaming\nltk_data...
[nltk_data]   Package vader_lexicon is already up-to-date!





  0%|          | 0/50000 [00:00<?, ?it/s]




  0%|          | 5/50000 [00:00<17:17, 48.21it/s]




  0%|          | 10/50000 [00:00<17:18, 48.13it/s]




  0%|          | 13/50000 [00:00<22:22, 37.23it/s]




  0%|          | 18/50000 [00:00<21:29, 38.75it/s]




  0%|          | 24/50000 [00:00<19:14, 43.29it/s]




  0%|          | 31/50000 [00:00<17:14, 48.31it/s]




  0%|          | 37/50000 [00:00<16:43, 49.81it/s]




  0%|          | 43/50000 [00:00<16:03, 51.84it/s]




  0%|          | 49/50000 [00:00<15:55, 52.25it/s]




  0%|          | 55/50000 [00:01<17:20, 48.02it/s]




  0%|          | 62/50000 [00:01<15:44, 52.85it/s]




  0%|          | 68/50000 [00:01<15:42, 52.96it/s]




  0%|          | 74/50000 [00:01<15:26, 53.90it/s]




  0%|          | 80/50000 [00:01<15:55, 52.26it/s]




  0%|          | 86/50000 [00:01<15:21, 54.18it/s]




  0%|          | 92/50000 [00:01<15:03, 55.23it/s]




  0%|          | 100/50000 [00:01<14:03, 59.17it/s]




  0%|          | 107/50000 [00:02<14:12, 58.52it/s]




  0%|          | 113/50000 [00:02<14:37, 56.83it/s]




  0%|          | 119/50000 [00:02<15:20, 54.20it/s]




  0%|          | 126/50000 [00:02<14:34, 57.04it/s]




  0%|          | 134/50000 [00:02<13:31, 61.45it/s]




  0%|          | 141/50000 [00:02<13:23, 62.05it/s]




  0%|          | 148/50000 [00:02<12:59, 63.94it/s]




  0%|          | 155/50000 [00:02<12:47, 64.96it/s]




  0%|          | 162/50000 [00:02<13:36, 61.07it/s]




  0%|          | 169/50000 [00:03<13:23, 62.03it/s]




  0%|          | 177/50000 [00:03<12:34, 66.06it/s]




  0%|          | 184/50000 [00:03<12:35, 65.92it/s]




  0%|          | 191/50000 [00:03<12:38, 65.64it/s]




  0%|          | 198/50000 [00:03<13:12, 62.81it/s]




  0%|          | 205/50000 [00:03<13:19, 62.26it/s]




  0%|          | 212/50000 [00:03<13:45, 60.29it/s]




  0%|          | 219/50000 [00:03<13:42, 60.51it/s]




  0%|          | 226/50000 [00:03<13:23, 61.95it/s]




  0%|          | 233/50000 [00:04<13:14, 62.67it/s]




  0%|          | 240/50000 [00:04<12:50, 64.57it/s]




  0%|          | 247/50000 [00:04<12:57, 63.99it/s]




  1%|          | 254/50000 [00:04<12:49, 64.64it/s]




  1%|          | 261/50000 [00:04<12:55, 64.12it/s]




  1%|          | 268/50000 [00:04<13:15, 62.49it/s]




  1%|          | 276/50000 [00:04<12:56, 64.04it/s]




  1%|          | 284/50000 [00:04<12:26, 66.63it/s]




  1%|          | 291/50000 [00:04<12:55, 64.14it/s]




  1%|          | 298/50000 [00:05<12:39, 65.47it/s]




  1%|          | 305/50000 [00:05<12:36, 65.69it/s]




  1%|          | 312/50000 [00:05<12:34, 65.85it/s]




  1%|          | 319/50000 [00:05<12:35, 65.77it/s]




  1%|          | 326/50000 [00:05<12:42, 65.17it/s]




  1%|          | 333/50000 [00:05<12:55, 64.05it/s]




  1%|          | 340/50000 [00:05<13:07, 63.02it/s]




  1%|          | 347/50000 [00:05<13:13, 62.58it/s]




  1%|          | 354/50000 [00:05<12:58, 63.80it/s]




  1%|          | 361/50000 [00:06<13:04, 63.28it/s]




  1%|          | 369/50000 [00:06<12:35, 65.73it/s]




  1%|          | 376/50000 [00:06<12:35, 65.69it/s]




  1%|          | 383/50000 [00:06<12:54, 64.05it/s]




  1%|          | 390/50000 [00:06<12:44, 64.86it/s]




  1%|          | 398/50000 [00:06<12:10, 67.94it/s]




  1%|          | 405/50000 [00:06<12:26, 66.45it/s]




  1%|          | 412/50000 [00:06<12:22, 66.76it/s]




  1%|          | 419/50000 [00:06<12:17, 67.26it/s]




  1%|          | 426/50000 [00:06<12:18, 67.14it/s]




  1%|          | 433/50000 [00:07<12:53, 64.11it/s]




  1%|          | 440/50000 [00:07<12:47, 64.55it/s]




  1%|          | 447/50000 [00:07<12:50, 64.32it/s]




  1%|          | 454/50000 [00:07<12:52, 64.16it/s]




  1%|          | 461/50000 [00:07<12:38, 65.31it/s]




  1%|          | 468/50000 [00:07<12:26, 66.32it/s]




  1%|          | 475/50000 [00:07<12:22, 66.67it/s]




  1%|          | 482/50000 [00:07<12:47, 64.51it/s]




  1%|          | 489/50000 [00:07<13:19, 61.92it/s]




  1%|          | 496/50000 [00:08<13:03, 63.15it/s]




  1%|          | 503/50000 [00:08<13:43, 60.09it/s]




  1%|          | 511/50000 [00:08<12:52, 64.05it/s]




  1%|          | 518/50000 [00:08<13:33, 60.81it/s]




  1%|          | 525/50000 [00:08<13:16, 62.09it/s]




  1%|          | 532/50000 [00:08<13:48, 59.72it/s]




  1%|          | 539/50000 [00:08<13:36, 60.58it/s]




  1%|          | 546/50000 [00:08<13:16, 62.10it/s]




  1%|          | 553/50000 [00:08<12:56, 63.70it/s]




  1%|          | 560/50000 [00:09<13:14, 62.21it/s]




  1%|          | 567/50000 [00:09<13:50, 59.49it/s]




  1%|          | 574/50000 [00:09<13:59, 58.89it/s]




  1%|          | 581/50000 [00:09<13:50, 59.52it/s]




  1%|          | 588/50000 [00:09<13:33, 60.74it/s]




  1%|          | 595/50000 [00:09<14:03, 58.55it/s]




  1%|          | 604/50000 [00:09<12:47, 64.32it/s]




  1%|          | 611/50000 [00:09<12:52, 63.90it/s]




  1%|          | 619/50000 [00:10<12:23, 66.44it/s]




  1%|▏         | 626/50000 [00:10<12:47, 64.37it/s]




  1%|▏         | 633/50000 [00:10<12:31, 65.68it/s]




  1%|▏         | 640/50000 [00:10<12:36, 65.29it/s]




  1%|▏         | 647/50000 [00:10<12:23, 66.34it/s]




  1%|▏         | 654/50000 [00:10<12:44, 64.56it/s]




  1%|▏         | 661/50000 [00:10<12:35, 65.32it/s]




  1%|▏         | 668/50000 [00:10<12:28, 65.94it/s]




  1%|▏         | 676/50000 [00:10<12:14, 67.19it/s]




  1%|▏         | 683/50000 [00:11<12:57, 63.40it/s]




  1%|▏         | 690/50000 [00:11<13:09, 62.46it/s]




  1%|▏         | 697/50000 [00:11<13:08, 62.52it/s]




  1%|▏         | 704/50000 [00:11<12:55, 63.58it/s]




  1%|▏         | 711/50000 [00:11<14:16, 57.55it/s]




  1%|▏         | 717/50000 [00:11<15:55, 51.56it/s]




  1%|▏         | 725/50000 [00:11<14:15, 57.62it/s]




  1%|▏         | 732/50000 [00:11<14:00, 58.61it/s]




  1%|▏         | 739/50000 [00:12<15:37, 52.52it/s]




  1%|▏         | 745/50000 [00:12<15:46, 52.05it/s]




  2%|▏         | 752/50000 [00:12<14:58, 54.84it/s]




  2%|▏         | 758/50000 [00:12<14:53, 55.09it/s]




  2%|▏         | 764/50000 [00:12<15:14, 53.86it/s]




  2%|▏         | 771/50000 [00:12<14:21, 57.12it/s]




  2%|▏         | 779/50000 [00:12<13:22, 61.30it/s]




  2%|▏         | 786/50000 [00:12<14:09, 57.96it/s]




  2%|▏         | 793/50000 [00:12<13:47, 59.45it/s]




  2%|▏         | 800/50000 [00:13<13:13, 61.98it/s]




  2%|▏         | 807/50000 [00:13<12:52, 63.71it/s]




  2%|▏         | 815/50000 [00:13<12:11, 67.21it/s]




  2%|▏         | 822/50000 [00:13<12:33, 65.23it/s]




  2%|▏         | 829/50000 [00:13<13:46, 59.52it/s]




  2%|▏         | 836/50000 [00:13<15:06, 54.26it/s]




  2%|▏         | 842/50000 [00:13<15:24, 53.16it/s]




  2%|▏         | 848/50000 [00:13<15:33, 52.63it/s]




  2%|▏         | 854/50000 [00:14<16:36, 49.31it/s]




  2%|▏         | 860/50000 [00:14<17:10, 47.67it/s]




  2%|▏         | 867/50000 [00:14<15:56, 51.39it/s]




  2%|▏         | 873/50000 [00:14<16:16, 50.29it/s]




  2%|▏         | 879/50000 [00:14<17:47, 46.03it/s]




  2%|▏         | 884/50000 [00:14<17:37, 46.47it/s]




  2%|▏         | 890/50000 [00:14<16:32, 49.50it/s]




  2%|▏         | 896/50000 [00:14<16:50, 48.61it/s]




  2%|▏         | 901/50000 [00:15<18:11, 44.96it/s]




  2%|▏         | 906/50000 [00:15<19:17, 42.40it/s]




  2%|▏         | 914/50000 [00:15<16:50, 48.57it/s]




  2%|▏         | 920/50000 [00:15<15:56, 51.29it/s]




  2%|▏         | 926/50000 [00:15<16:07, 50.74it/s]




  2%|▏         | 933/50000 [00:15<15:18, 53.44it/s]




  2%|▏         | 940/50000 [00:15<14:29, 56.45it/s]




  2%|▏         | 947/50000 [00:15<13:48, 59.21it/s]



 30%|██▉       | 14971/50000 [04:01<09:22, 62.31it/s]




  2%|▏         | 954/50000 [00:15<13:57, 58.55it/s]




  2%|▏         | 960/50000 [00:16<14:21, 56.93it/s]




  2%|▏         | 966/50000 [00:16<15:09, 53.89it/s]




  2%|▏         | 972/50000 [00:16<18:08, 45.06it/s]




  2%|▏         | 981/50000 [00:16<15:38, 52.21it/s]




  2%|▏         | 987/50000 [00:16<17:02, 47.93it/s]




  2%|▏         | 993/50000 [00:16<17:13, 47.42it/s]




  2%|▏         | 999/50000 [00:16<18:09, 44.96it/s]




  2%|▏         | 1004/50000 [00:17<18:58, 43.05it/s]




  2%|▏         | 1009/50000 [00:17<18:44, 43.55it/s]




  2%|▏         | 1015/50000 [00:17<17:58, 45.44it/s]




  2%|▏         | 1022/50000 [00:17<16:41, 48.90it/s]




  2%|▏         | 1030/50000 [00:17<15:17, 53.40it/s]




  2%|▏         | 1037/50000 [00:17<14:53, 54.83it/s]




  2%|▏         | 1043/50000 [00:17<15:03, 54.20it/s]




  2%|▏         | 1051/50000 [00:17<13:46, 59.24it/s]




  2%|▏         | 1058/50000 [00:17<14:11, 57.49it/s]




  2%|▏         | 1064/50000 [00:18<14:34, 55.99it/s]




  2%|▏         | 1070/50000 [00:18<14:45, 55.29it/s]




  2%|▏         | 1077/50000 [00:18<13:54, 58.60it/s]




  2%|▏         | 1083/50000 [00:18<14:17, 57.05it/s]




  2%|▏         | 1091/50000 [00:18<13:12, 61.74it/s]




  2%|▏         | 1098/50000 [00:18<14:02, 58.01it/s]




  2%|▏         | 1105/50000 [00:18<13:37, 59.79it/s]




  2%|▏         | 1113/50000 [00:18<12:43, 64.04it/s]




  2%|▏         | 1120/50000 [00:18<12:50, 63.45it/s]




  2%|▏         | 1127/50000 [00:19<12:54, 63.12it/s]




  2%|▏         | 1134/50000 [00:19<12:55, 62.98it/s]




  2%|▏         | 1141/50000 [00:19<12:46, 63.74it/s]




  2%|▏         | 1148/50000 [00:19<12:43, 64.02it/s]




  2%|▏         | 1155/50000 [00:19<12:39, 64.31it/s]




  2%|▏         | 1162/50000 [00:19<12:34, 64.69it/s]




  2%|▏         | 1169/50000 [00:19<12:38, 64.42it/s]




  2%|▏         | 1176/50000 [00:19<12:35, 64.59it/s]




  2%|▏         | 1183/50000 [00:19<12:59, 62.63it/s]




  2%|▏         | 1190/50000 [00:20<12:38, 64.36it/s]




  2%|▏         | 1199/50000 [00:20<11:49, 68.81it/s]




  2%|▏         | 1206/50000 [00:20<13:52, 58.64it/s]




  2%|▏         | 1214/50000 [00:20<13:01, 62.46it/s]




  2%|▏         | 1221/50000 [00:20<12:56, 62.86it/s]




  2%|▏         | 1228/50000 [00:20<13:04, 62.13it/s]




  2%|▏         | 1236/50000 [00:20<12:18, 66.05it/s]




  2%|▏         | 1243/50000 [00:20<12:28, 65.18it/s]




  2%|▎         | 1250/50000 [00:21<12:31, 64.85it/s]




  3%|▎         | 1257/50000 [00:21<12:36, 64.44it/s]




  3%|▎         | 1264/50000 [00:21<12:47, 63.46it/s]




  3%|▎         | 1271/50000 [00:21<12:48, 63.39it/s]




  3%|▎         | 1278/50000 [00:21<12:42, 63.86it/s]




  3%|▎         | 1285/50000 [00:21<12:49, 63.33it/s]




  3%|▎         | 1292/50000 [00:21<13:04, 62.12it/s]




  3%|▎         | 1300/50000 [00:21<12:27, 65.17it/s]




  3%|▎         | 1307/50000 [00:21<12:19, 65.85it/s]




  3%|▎         | 1314/50000 [00:22<12:26, 65.22it/s]




  3%|▎         | 1321/50000 [00:22<12:26, 65.24it/s]




  3%|▎         | 1329/50000 [00:22<12:02, 67.37it/s]




  3%|▎         | 1336/50000 [00:22<13:04, 62.04it/s]




  3%|▎         | 1344/50000 [00:22<12:19, 65.83it/s]




  3%|▎         | 1351/50000 [00:22<12:37, 64.19it/s]




  3%|▎         | 1358/50000 [00:22<12:36, 64.33it/s]




  3%|▎         | 1365/50000 [00:22<12:48, 63.31it/s]




  3%|▎         | 1372/50000 [00:22<13:11, 61.46it/s]




  3%|▎         | 1379/50000 [00:23<13:19, 60.85it/s]




  3%|▎         | 1386/50000 [00:23<12:57, 62.53it/s]




  3%|▎         | 1393/50000 [00:23<12:46, 63.42it/s]




  3%|▎         | 1400/50000 [00:23<12:53, 62.85it/s]




  3%|▎         | 1407/50000 [00:23<12:47, 63.31it/s]




  3%|▎         | 1415/50000 [00:23<12:20, 65.59it/s]




  3%|▎         | 1422/50000 [00:23<12:48, 63.20it/s]




  3%|▎         | 1430/50000 [00:23<12:14, 66.16it/s]




  3%|▎         | 1438/50000 [00:23<11:44, 68.93it/s]




  3%|▎         | 1446/50000 [00:24<11:27, 70.63it/s]




  3%|▎         | 1454/50000 [00:24<11:53, 68.04it/s]




  3%|▎         | 1461/50000 [00:24<12:11, 66.33it/s]




  3%|▎         | 1468/50000 [00:24<12:03, 67.05it/s]




  3%|▎         | 1476/50000 [00:24<11:57, 67.67it/s]




  3%|▎         | 1483/50000 [00:24<12:09, 66.46it/s]




  3%|▎         | 1491/50000 [00:24<11:45, 68.80it/s]




  3%|▎         | 1499/50000 [00:24<11:40, 69.26it/s]




  3%|▎         | 1506/50000 [00:24<12:14, 66.01it/s]




  3%|▎         | 1513/50000 [00:25<12:44, 63.38it/s]




  3%|▎         | 1522/50000 [00:25<11:42, 68.96it/s]




  3%|▎         | 1530/50000 [00:25<12:26, 64.91it/s]




  3%|▎         | 1537/50000 [00:25<12:11, 66.22it/s]




  3%|▎         | 1546/50000 [00:25<11:47, 68.52it/s]




  3%|▎         | 1553/50000 [00:25<12:27, 64.81it/s]




  3%|▎         | 1560/50000 [00:25<12:29, 64.66it/s]




  3%|▎         | 1567/50000 [00:25<12:50, 62.85it/s]




  3%|▎         | 1574/50000 [00:25<13:28, 59.90it/s]




  3%|▎         | 1581/50000 [00:26<14:40, 55.00it/s]




  3%|▎         | 1587/50000 [00:26<14:27, 55.82it/s]




  3%|▎         | 1594/50000 [00:26<13:35, 59.32it/s]




  3%|▎         | 1601/50000 [00:26<13:22, 60.29it/s]




  3%|▎         | 1608/50000 [00:26<14:07, 57.11it/s]




  3%|▎         | 1615/50000 [00:26<13:26, 60.03it/s]




  3%|▎         | 1622/50000 [00:26<13:21, 60.38it/s]




  3%|▎         | 1629/50000 [00:26<13:53, 58.02it/s]




  3%|▎         | 1637/50000 [00:27<12:57, 62.24it/s]




  3%|▎         | 1644/50000 [00:27<13:43, 58.69it/s]




  3%|▎         | 1651/50000 [00:27<13:54, 57.90it/s]




  3%|▎         | 1657/50000 [00:27<14:31, 55.48it/s]




  3%|▎         | 1663/50000 [00:27<14:37, 55.09it/s]




  3%|▎         | 1669/50000 [00:27<14:34, 55.27it/s]




  3%|▎         | 1675/50000 [00:27<15:25, 52.23it/s]




  3%|▎         | 1681/50000 [00:27<15:05, 53.37it/s]




  3%|▎         | 1689/50000 [00:28<13:54, 57.91it/s]




  3%|▎         | 1696/50000 [00:28<13:35, 59.26it/s]




  3%|▎         | 1703/50000 [00:28<13:13, 60.87it/s]




  3%|▎         | 1710/50000 [00:28<13:31, 59.52it/s]




  3%|▎         | 1717/50000 [00:28<13:31, 59.51it/s]




  3%|▎         | 1725/50000 [00:28<12:46, 62.99it/s]




  3%|▎         | 1732/50000 [00:28<12:56, 62.14it/s]




  3%|▎         | 1739/50000 [00:28<13:01, 61.72it/s]




  3%|▎         | 1746/50000 [00:28<13:37, 59.03it/s]




  4%|▎         | 1752/50000 [00:29<14:18, 56.20it/s]




  4%|▎         | 1759/50000 [00:29<13:55, 57.71it/s]




  4%|▎         | 1767/50000 [00:29<12:59, 61.85it/s]




  4%|▎         | 1774/50000 [00:29<12:58, 61.93it/s]




  4%|▎         | 1781/50000 [00:29<13:32, 59.31it/s]




  4%|▎         | 1788/50000 [00:29<15:00, 53.52it/s]




  4%|▎         | 1794/50000 [00:29<17:33, 45.76it/s]




  4%|▎         | 1799/50000 [00:30<20:47, 38.62it/s]




  4%|▎         | 1804/50000 [00:30<21:18, 37.71it/s]




  4%|▎         | 1809/50000 [00:30<21:43, 36.98it/s]




  4%|▎         | 1813/50000 [00:30<21:23, 37.54it/s]




  4%|▎         | 1818/50000 [00:30<21:07, 38.02it/s]




  4%|▎         | 1823/50000 [00:30<19:44, 40.69it/s]




  4%|▎         | 1828/50000 [00:30<18:39, 43.01it/s]




  4%|▎         | 1833/50000 [00:30<18:41, 42.96it/s]




  4%|▎         | 1839/50000 [00:30<17:38, 45.50it/s]




  4%|▎         | 1845/50000 [00:31<16:55, 47.40it/s]




  4%|▎         | 1850/50000 [00:31<17:23, 46.13it/s]




  4%|▎         | 1855/50000 [00:31<17:14, 46.53it/s]




  4%|▎         | 1860/50000 [00:31<17:00, 47.16it/s]




  4%|▎         | 1866/50000 [00:31<16:54, 47.44it/s]




  4%|▎         | 1872/50000 [00:31<15:52, 50.53it/s]




  4%|▍         | 1879/50000 [00:31<14:57, 53.65it/s]




  4%|▍         | 1886/50000 [00:31<14:07, 56.75it/s]




  4%|▍         | 1892/50000 [00:31<14:53, 53.85it/s]




  4%|▍         | 1898/50000 [00:32<14:49, 54.10it/s]




  4%|▍         | 1904/50000 [00:32<15:34, 51.49it/s]




  4%|▍         | 1910/50000 [00:32<15:55, 50.31it/s]




  4%|▍         | 1916/50000 [00:32<16:24, 48.87it/s]




  4%|▍         | 1924/50000 [00:32<14:42, 54.48it/s]




  4%|▍         | 1930/50000 [00:32<15:39, 51.17it/s]




  4%|▍         | 1936/50000 [00:32<15:55, 50.28it/s]




  4%|▍         | 1942/50000 [00:32<16:15, 49.25it/s]




  4%|▍         | 1948/50000 [00:33<17:41, 45.26it/s]




  4%|▍         | 1953/50000 [00:33<18:39, 42.91it/s]




  4%|▍         | 1960/50000 [00:33<16:45, 47.78it/s]




  4%|▍         | 1966/50000 [00:33<15:50, 50.54it/s]




  4%|▍         | 1972/50000 [00:33<16:47, 47.66it/s]




  4%|▍         | 1979/50000 [00:33<15:33, 51.46it/s]




  4%|▍         | 1985/50000 [00:33<15:39, 51.12it/s]




  4%|▍         | 1991/50000 [00:33<15:56, 50.18it/s]




  4%|▍         | 1997/50000 [00:34<15:42, 50.93it/s]




  4%|▍         | 2005/50000 [00:34<14:02, 56.95it/s]




  4%|▍         | 2012/50000 [00:34<13:16, 60.21it/s]




  4%|▍         | 2020/50000 [00:34<12:24, 64.43it/s]




  4%|▍         | 2027/50000 [00:34<12:16, 65.13it/s]




  4%|▍         | 2034/50000 [00:34<12:45, 62.65it/s]




  4%|▍         | 2041/50000 [00:34<12:29, 64.03it/s]




  4%|▍         | 2049/50000 [00:34<12:03, 66.29it/s]




  4%|▍         | 2056/50000 [00:34<12:09, 65.71it/s]




  4%|▍         | 2063/50000 [00:35<12:05, 66.05it/s]




  4%|▍         | 2070/50000 [00:35<12:15, 65.18it/s]




  4%|▍         | 2077/50000 [00:35<12:11, 65.48it/s]




  4%|▍         | 2084/50000 [00:35<12:19, 64.80it/s]




  4%|▍         | 2092/50000 [00:35<11:44, 67.97it/s]




  4%|▍         | 2099/50000 [00:35<12:53, 61.90it/s]




  4%|▍         | 2106/50000 [00:35<13:11, 60.52it/s]




  4%|▍         | 2113/50000 [00:35<12:54, 61.85it/s]




  4%|▍         | 2120/50000 [00:35<12:29, 63.87it/s]




  4%|▍         | 2127/50000 [00:36<12:26, 64.11it/s]




  4%|▍         | 2135/50000 [00:36<11:55, 66.94it/s]




  4%|▍         | 2143/50000 [00:36<11:34, 68.89it/s]




  4%|▍         | 2151/50000 [00:36<11:24, 69.87it/s]




  4%|▍         | 2159/50000 [00:36<11:25, 69.83it/s]




  4%|▍         | 2167/50000 [00:36<11:28, 69.45it/s]




  4%|▍         | 2174/50000 [00:36<11:44, 67.85it/s]




  4%|▍         | 2181/50000 [00:36<11:47, 67.54it/s]




  4%|▍         | 2188/50000 [00:36<12:14, 65.09it/s]




  4%|▍         | 2195/50000 [00:37<12:41, 62.79it/s]




  4%|▍         | 2203/50000 [00:37<12:20, 64.57it/s]




  4%|▍         | 2210/50000 [00:37<12:16, 64.87it/s]




  4%|▍         | 2217/50000 [00:37<12:06, 65.76it/s]




  4%|▍         | 2224/50000 [00:37<12:36, 63.14it/s]




  4%|▍         | 2232/50000 [00:37<12:06, 65.78it/s]




  4%|▍         | 2240/50000 [00:37<11:50, 67.26it/s]




  4%|▍         | 2247/50000 [00:37<12:05, 65.82it/s]




  5%|▍         | 2254/50000 [00:37<12:14, 65.02it/s]




  5%|▍         | 2261/50000 [00:38<12:26, 63.94it/s]




  5%|▍         | 2268/50000 [00:38<12:57, 61.39it/s]




  5%|▍         | 2275/50000 [00:38<12:52, 61.77it/s]




  5%|▍         | 2283/50000 [00:38<12:15, 64.89it/s]




  5%|▍         | 2290/50000 [00:38<12:23, 64.21it/s]




  5%|▍         | 2298/50000 [00:38<12:08, 65.52it/s]




  5%|▍         | 2305/50000 [00:38<12:03, 65.91it/s]




  5%|▍         | 2312/50000 [00:38<12:18, 64.55it/s]




  5%|▍         | 2319/50000 [00:38<12:05, 65.68it/s]




  5%|▍         | 2327/50000 [00:39<11:45, 67.53it/s]




  5%|▍         | 2334/50000 [00:39<12:00, 66.18it/s]




  5%|▍         | 2343/50000 [00:39<11:19, 70.09it/s]




  5%|▍         | 2351/50000 [00:39<11:01, 72.06it/s]




  5%|▍         | 2359/50000 [00:39<11:02, 71.93it/s]




  5%|▍         | 2367/50000 [00:39<11:10, 71.07it/s]




  5%|▍         | 2375/50000 [00:39<11:04, 71.62it/s]




  5%|▍         | 2383/50000 [00:39<11:37, 68.23it/s]




  5%|▍         | 2391/50000 [00:39<11:16, 70.40it/s]




  5%|▍         | 2399/50000 [00:40<11:20, 69.93it/s]




  5%|▍         | 2407/50000 [00:40<11:24, 69.51it/s]




  5%|▍         | 2414/50000 [00:40<12:35, 62.96it/s]




  5%|▍         | 2421/50000 [00:40<12:55, 61.39it/s]




  5%|▍         | 2429/50000 [00:40<12:08, 65.31it/s]




  5%|▍         | 2437/50000 [00:40<11:28, 69.08it/s]




  5%|▍         | 2445/50000 [00:40<11:18, 70.09it/s]




  5%|▍         | 2453/50000 [00:40<11:37, 68.20it/s]




  5%|▍         | 2460/50000 [00:40<11:35, 68.39it/s]




  5%|▍         | 2467/50000 [00:41<12:36, 62.81it/s]




  5%|▍         | 2474/50000 [00:41<13:05, 60.49it/s]




  5%|▍         | 2481/50000 [00:41<13:15, 59.73it/s]




  5%|▍         | 2488/50000 [00:41<12:52, 61.54it/s]




  5%|▍         | 2495/50000 [00:41<14:22, 55.09it/s]




  5%|▌         | 2501/50000 [00:41<15:18, 51.71it/s]




  5%|▌         | 2507/50000 [00:41<14:42, 53.83it/s]




  5%|▌         | 2513/50000 [00:42<15:30, 51.05it/s]




  5%|▌         | 2519/50000 [00:42<15:11, 52.08it/s]




  5%|▌         | 2525/50000 [00:42<14:53, 53.12it/s]




  5%|▌         | 2532/50000 [00:42<13:57, 56.69it/s]




  5%|▌         | 2538/50000 [00:42<14:49, 53.37it/s]




  5%|▌         | 2544/50000 [00:42<14:42, 53.76it/s]




  5%|▌         | 2551/50000 [00:42<14:11, 55.75it/s]




  5%|▌         | 2557/50000 [00:42<13:54, 56.85it/s]




  5%|▌         | 2564/50000 [00:42<13:12, 59.82it/s]




  5%|▌         | 2571/50000 [00:42<12:47, 61.77it/s]




  5%|▌         | 2578/50000 [00:43<12:38, 62.53it/s]




  5%|▌         | 2585/50000 [00:43<12:49, 61.58it/s]




  5%|▌         | 2592/50000 [00:43<12:49, 61.58it/s]




  5%|▌         | 2599/50000 [00:43<13:28, 58.64it/s]




  5%|▌         | 2605/50000 [00:43<14:11, 55.65it/s]




  5%|▌         | 2611/50000 [00:43<15:17, 51.65it/s]




  5%|▌         | 2617/50000 [00:43<15:31, 50.85it/s]




  5%|▌         | 2623/50000 [00:43<16:31, 47.80it/s]




  5%|▌         | 2628/50000 [00:44<17:06, 46.13it/s]




  5%|▌         | 2633/50000 [00:44<18:10, 43.45it/s]




  5%|▌         | 2638/50000 [00:44<18:11, 43.38it/s]




  5%|▌         | 2643/50000 [00:44<18:44, 42.13it/s]




  5%|▌         | 2648/50000 [00:44<19:29, 40.50it/s]




  5%|▌         | 2653/50000 [00:44<18:30, 42.65it/s]




  5%|▌         | 2658/50000 [00:44<17:55, 44.00it/s]




  5%|▌         | 2663/50000 [00:44<17:33, 44.94it/s]




  5%|▌         | 2668/50000 [00:45<17:20, 45.50it/s]




  5%|▌         | 2673/50000 [00:45<17:35, 44.85it/s]




  5%|▌         | 2679/50000 [00:45<17:12, 45.83it/s]




  5%|▌         | 2684/50000 [00:45<17:08, 46.01it/s]




  5%|▌         | 2689/50000 [00:45<19:12, 41.03it/s]




  5%|▌         | 2694/50000 [00:45<18:27, 42.73it/s]




  5%|▌         | 2699/50000 [00:45<18:54, 41.70it/s]




  5%|▌         | 2704/50000 [00:45<19:04, 41.31it/s]




  5%|▌         | 2709/50000 [00:45<18:42, 42.13it/s]




  5%|▌         | 2714/50000 [00:46<18:55, 41.65it/s]




  5%|▌         | 2719/50000 [00:46<19:19, 40.78it/s]




  5%|▌         | 2724/50000 [00:46<19:08, 41.17it/s]




  5%|▌         | 2729/50000 [00:46<19:11, 41.05it/s]




  5%|▌         | 2734/50000 [00:46<19:47, 39.79it/s]




  5%|▌         | 2739/50000 [00:46<19:11, 41.06it/s]




  5%|▌         | 2745/50000 [00:46<17:42, 44.48it/s]




  6%|▌         | 2750/50000 [00:46<18:17, 43.07it/s]




  6%|▌         | 2755/50000 [00:47<19:57, 39.45it/s]




  6%|▌         | 2760/50000 [00:47<19:43, 39.93it/s]




  6%|▌         | 2766/50000 [00:47<18:00, 43.72it/s]




  6%|▌         | 2771/50000 [00:47<17:32, 44.86it/s]




  6%|▌         | 2777/50000 [00:47<16:21, 48.10it/s]




  6%|▌         | 2783/50000 [00:47<15:23, 51.11it/s]




  6%|▌         | 2789/50000 [00:47<14:59, 52.47it/s]




  6%|▌         | 2795/50000 [00:47<14:45, 53.33it/s]




  6%|▌         | 2801/50000 [00:47<15:04, 52.19it/s]




  6%|▌         | 2808/50000 [00:48<14:14, 55.21it/s]




  6%|▌         | 2815/50000 [00:48<13:42, 57.39it/s]




  6%|▌         | 2821/50000 [00:48<13:49, 56.87it/s]




  6%|▌         | 2827/50000 [00:48<14:25, 54.50it/s]




  6%|▌         | 2833/50000 [00:48<14:41, 53.54it/s]




  6%|▌         | 2839/50000 [00:48<16:39, 47.16it/s]




  6%|▌         | 2844/50000 [00:48<16:40, 47.13it/s]




  6%|▌         | 2849/50000 [00:48<17:51, 44.01it/s]




  6%|▌         | 2854/50000 [00:49<17:28, 44.95it/s]




  6%|▌         | 2860/50000 [00:49<16:35, 47.37it/s]




  6%|▌         | 2866/50000 [00:49<16:07, 48.69it/s]




  6%|▌         | 2875/50000 [00:49<14:04, 55.77it/s]




  6%|▌         | 2881/50000 [00:49<14:22, 54.63it/s]




  6%|▌         | 2887/50000 [00:49<14:32, 53.98it/s]




  6%|▌         | 2893/50000 [00:49<14:16, 55.01it/s]




  6%|▌         | 2899/50000 [00:49<14:13, 55.21it/s]




  6%|▌         | 2905/50000 [00:49<13:56, 56.29it/s]




  6%|▌         | 2912/50000 [00:50<13:16, 59.09it/s]




  6%|▌         | 2920/50000 [00:50<12:45, 61.50it/s]




  6%|▌         | 2927/50000 [00:50<12:57, 60.57it/s]




  6%|▌         | 2934/50000 [00:50<12:37, 62.16it/s]




  6%|▌         | 2941/50000 [00:50<12:23, 63.32it/s]




  6%|▌         | 2948/50000 [00:50<12:04, 64.96it/s]




  6%|▌         | 2956/50000 [00:50<11:27, 68.45it/s]




  6%|▌         | 2963/50000 [00:50<11:36, 67.57it/s]




  6%|▌         | 2970/50000 [00:50<12:22, 63.34it/s]




  6%|▌         | 2977/50000 [00:51<12:20, 63.54it/s]




  6%|▌         | 2984/50000 [00:51<12:03, 64.98it/s]




  6%|▌         | 2991/50000 [00:51<12:18, 63.66it/s]




  6%|▌         | 2998/50000 [00:51<12:18, 63.61it/s]




  6%|▌         | 3005/50000 [00:51<12:21, 63.41it/s]




  6%|▌         | 3012/50000 [00:51<12:20, 63.44it/s]




  6%|▌         | 3019/50000 [00:51<12:21, 63.38it/s]




  6%|▌         | 3026/50000 [00:51<12:51, 60.86it/s]




  6%|▌         | 3033/50000 [00:51<12:39, 61.88it/s]




  6%|▌         | 3040/50000 [00:52<13:08, 59.58it/s]




  6%|▌         | 3047/50000 [00:52<12:46, 61.27it/s]




  6%|▌         | 3055/50000 [00:52<12:09, 64.35it/s]




  6%|▌         | 3062/50000 [00:52<12:11, 64.19it/s]




  6%|▌         | 3069/50000 [00:52<12:39, 61.79it/s]




  6%|▌         | 3076/50000 [00:52<13:46, 56.78it/s]




  6%|▌         | 3082/50000 [00:52<14:16, 54.76it/s]




  6%|▌         | 3088/50000 [00:52<14:15, 54.85it/s]




  6%|▌         | 3096/50000 [00:52<13:03, 59.85it/s]




  6%|▌         | 3103/50000 [00:53<13:18, 58.76it/s]




  6%|▌         | 3111/50000 [00:53<12:34, 62.11it/s]




  6%|▌         | 3118/50000 [00:53<12:36, 61.94it/s]




  6%|▋         | 3125/50000 [00:53<12:42, 61.50it/s]




  6%|▋         | 3132/50000 [00:53<12:17, 63.53it/s]




  6%|▋         | 3139/50000 [00:53<12:00, 65.02it/s]




  6%|▋         | 3146/50000 [00:53<11:46, 66.31it/s]




  6%|▋         | 3154/50000 [00:53<11:25, 68.33it/s]




  6%|▋         | 3162/50000 [00:53<10:59, 71.04it/s]




  6%|▋         | 3170/50000 [00:54<10:52, 71.78it/s]




  6%|▋         | 3178/50000 [00:54<11:33, 67.55it/s]




  6%|▋         | 3185/50000 [00:54<11:48, 66.09it/s]




  6%|▋         | 3192/50000 [00:54<12:03, 64.66it/s]




  6%|▋         | 3199/50000 [00:54<12:45, 61.14it/s]




  6%|▋         | 3206/50000 [00:54<12:45, 61.11it/s]




  6%|▋         | 3214/50000 [00:54<11:57, 65.17it/s]




  6%|▋         | 3221/50000 [00:54<12:04, 64.57it/s]




  6%|▋         | 3229/50000 [00:55<11:44, 66.37it/s]




  6%|▋         | 3236/50000 [00:55<11:37, 67.09it/s]




  6%|▋         | 3243/50000 [00:55<11:53, 65.51it/s]




  7%|▋         | 3251/50000 [00:55<11:35, 67.23it/s]




  7%|▋         | 3258/50000 [00:55<12:02, 64.70it/s]




  7%|▋         | 3266/50000 [00:55<11:25, 68.16it/s]




  7%|▋         | 3273/50000 [00:55<11:53, 65.49it/s]




  7%|▋         | 3280/50000 [00:55<11:53, 65.52it/s]




  7%|▋         | 3288/50000 [00:55<11:23, 68.38it/s]




  7%|▋         | 3297/50000 [00:55<10:41, 72.76it/s]




  7%|▋         | 3305/50000 [00:56<11:18, 68.78it/s]




  7%|▋         | 3313/50000 [00:56<11:35, 67.16it/s]




  7%|▋         | 3320/50000 [00:56<11:36, 67.06it/s]




  7%|▋         | 3327/50000 [00:56<11:48, 65.87it/s]




  7%|▋         | 3334/50000 [00:56<11:39, 66.72it/s]




  7%|▋         | 3341/50000 [00:56<12:08, 64.02it/s]




  7%|▋         | 3349/50000 [00:56<11:33, 67.29it/s]




  7%|▋         | 3356/50000 [00:56<12:06, 64.21it/s]




  7%|▋         | 3363/50000 [00:57<12:21, 62.88it/s]




  7%|▋         | 3370/50000 [00:57<12:04, 64.37it/s]




  7%|▋         | 3377/50000 [00:57<12:04, 64.38it/s]




  7%|▋         | 3385/50000 [00:57<12:01, 64.58it/s]




  7%|▋         | 3392/50000 [00:57<12:04, 64.34it/s]




  7%|▋         | 3399/50000 [00:57<11:52, 65.44it/s]




  7%|▋         | 3406/50000 [00:57<11:53, 65.30it/s]




  7%|▋         | 3413/50000 [00:57<12:22, 62.76it/s]




  7%|▋         | 3420/50000 [00:57<12:48, 60.62it/s]




  7%|▋         | 3427/50000 [00:58<12:20, 62.86it/s]




  7%|▋         | 3434/50000 [00:58<12:17, 63.14it/s]




  7%|▋         | 3441/50000 [00:58<12:21, 62.83it/s]




  7%|▋         | 3448/50000 [00:58<12:05, 64.16it/s]




  7%|▋         | 3455/50000 [00:58<12:05, 64.14it/s]




  7%|▋         | 3462/50000 [00:58<12:15, 63.31it/s]




  7%|▋         | 3469/50000 [00:58<12:17, 63.12it/s]




  7%|▋         | 3476/50000 [00:58<12:10, 63.67it/s]




  7%|▋         | 3483/50000 [00:58<12:32, 61.86it/s]




  7%|▋         | 3490/50000 [00:59<12:17, 63.03it/s]




  7%|▋         | 3497/50000 [00:59<12:20, 62.83it/s]




  7%|▋         | 3504/50000 [00:59<11:59, 64.63it/s]




  7%|▋         | 3511/50000 [00:59<12:05, 64.05it/s]




  7%|▋         | 3518/50000 [00:59<11:51, 65.31it/s]




  7%|▋         | 3527/50000 [00:59<10:57, 70.71it/s]




  7%|▋         | 3535/50000 [00:59<10:54, 70.98it/s]




  7%|▋         | 3543/50000 [00:59<11:32, 67.07it/s]




  7%|▋         | 3550/50000 [00:59<12:01, 64.42it/s]




  7%|▋         | 3559/50000 [01:00<11:08, 69.49it/s]




  7%|▋         | 3567/50000 [01:00<11:19, 68.32it/s]




  7%|▋         | 3574/50000 [01:00<11:31, 67.10it/s]




  7%|▋         | 3581/50000 [01:00<12:06, 63.91it/s]




  7%|▋         | 3588/50000 [01:00<11:54, 64.96it/s]




  7%|▋         | 3595/50000 [01:00<11:50, 65.33it/s]




  7%|▋         | 3602/50000 [01:00<12:04, 64.05it/s]




  7%|▋         | 3609/50000 [01:00<12:17, 62.94it/s]




  7%|▋         | 3616/50000 [01:00<13:00, 59.43it/s]




  7%|▋         | 3623/50000 [01:01<13:13, 58.47it/s]




  7%|▋         | 3630/50000 [01:01<12:41, 60.91it/s]




  7%|▋         | 3637/50000 [01:01<12:48, 60.32it/s]




  7%|▋         | 3644/50000 [01:01<12:20, 62.64it/s]




  7%|▋         | 3651/50000 [01:01<12:53, 59.92it/s]




  7%|▋         | 3658/50000 [01:01<13:03, 59.18it/s]




  7%|▋         | 3664/50000 [01:01<14:19, 53.88it/s]




  7%|▋         | 3670/50000 [01:01<15:03, 51.28it/s]




  7%|▋         | 3676/50000 [01:02<15:26, 49.98it/s]




  7%|▋         | 3682/50000 [01:02<16:07, 47.88it/s]




  7%|▋         | 3688/50000 [01:02<15:12, 50.74it/s]




  7%|▋         | 3694/50000 [01:02<14:45, 52.27it/s]




  7%|▋         | 3700/50000 [01:02<14:31, 53.11it/s]




  7%|▋         | 3706/50000 [01:02<15:50, 48.70it/s]




  7%|▋         | 3712/50000 [01:02<15:34, 49.54it/s]




  7%|▋         | 3718/50000 [01:02<16:05, 47.94it/s]




  7%|▋         | 3723/50000 [01:02<15:58, 48.29it/s]




  7%|▋         | 3728/50000 [01:03<16:48, 45.88it/s]




  7%|▋         | 3735/50000 [01:03<15:23, 50.10it/s]




  7%|▋         | 3741/50000 [01:03<14:43, 52.34it/s]




  7%|▋         | 3747/50000 [01:03<15:06, 51.00it/s]




  8%|▊         | 3754/50000 [01:03<14:06, 54.65it/s]




  8%|▊         | 3762/50000 [01:03<13:06, 58.76it/s]




  8%|▊         | 3769/50000 [01:03<13:04, 58.92it/s]




  8%|▊         | 3777/50000 [01:03<12:24, 62.08it/s]




  8%|▊         | 3784/50000 [01:04<12:34, 61.28it/s]




  8%|▊         | 3791/50000 [01:04<13:21, 57.62it/s]




  8%|▊         | 3797/50000 [01:04<14:41, 52.44it/s]




  8%|▊         | 3804/50000 [01:04<13:51, 55.58it/s]




  8%|▊         | 3810/50000 [01:04<14:33, 52.90it/s]




  8%|▊         | 3816/50000 [01:04<15:24, 49.97it/s]




  8%|▊         | 3822/50000 [01:04<16:34, 46.43it/s]




  8%|▊         | 3827/50000 [01:04<16:59, 45.29it/s]




  8%|▊         | 3833/50000 [01:05<16:38, 46.21it/s]




  8%|▊         | 3839/50000 [01:05<15:52, 48.47it/s]




  8%|▊         | 3846/50000 [01:05<14:47, 52.00it/s]




  8%|▊         | 3852/50000 [01:05<14:50, 51.83it/s]




  8%|▊         | 3859/50000 [01:05<14:04, 54.66it/s]




  8%|▊         | 3866/50000 [01:05<13:31, 56.84it/s]




  8%|▊         | 3873/50000 [01:05<12:49, 59.97it/s]




  8%|▊         | 3881/50000 [01:05<11:59, 64.13it/s]




  8%|▊         | 3889/50000 [01:05<11:18, 67.97it/s]




  8%|▊         | 3896/50000 [01:06<11:57, 64.28it/s]




  8%|▊         | 3903/50000 [01:06<12:00, 63.96it/s]




  8%|▊         | 3910/50000 [01:06<12:42, 60.45it/s]




  8%|▊         | 3917/50000 [01:06<12:32, 61.26it/s]




  8%|▊         | 3925/50000 [01:06<11:54, 64.50it/s]




  8%|▊         | 3932/50000 [01:06<11:50, 64.82it/s]




  8%|▊         | 3940/50000 [01:06<11:33, 66.39it/s]




  8%|▊         | 3947/50000 [01:06<11:26, 67.10it/s]




  8%|▊         | 3954/50000 [01:06<11:34, 66.26it/s]




  8%|▊         | 3961/50000 [01:07<11:34, 66.25it/s]




  8%|▊         | 3969/50000 [01:07<11:10, 68.64it/s]




  8%|▊         | 3976/50000 [01:07<11:53, 64.52it/s]




  8%|▊         | 3983/50000 [01:07<11:39, 65.75it/s]




  8%|▊         | 3991/50000 [01:07<11:21, 67.49it/s]




  8%|▊         | 3998/50000 [01:07<11:58, 63.98it/s]




  8%|▊         | 4005/50000 [01:07<12:13, 62.73it/s]




  8%|▊         | 4012/50000 [01:07<12:54, 59.37it/s]




  8%|▊         | 4019/50000 [01:07<13:05, 58.51it/s]




  8%|▊         | 4027/50000 [01:08<12:13, 62.64it/s]




  8%|▊         | 4034/50000 [01:08<12:15, 62.48it/s]




  8%|▊         | 4041/50000 [01:08<13:27, 56.91it/s]




  8%|▊         | 4047/50000 [01:08<13:25, 57.02it/s]




  8%|▊         | 4054/50000 [01:08<12:50, 59.65it/s]




  8%|▊         | 4061/50000 [01:08<12:23, 61.80it/s]




  8%|▊         | 4068/50000 [01:08<12:02, 63.58it/s]




  8%|▊         | 4075/50000 [01:08<12:14, 62.54it/s]




  8%|▊         | 4084/50000 [01:08<11:23, 67.19it/s]




  8%|▊         | 4091/50000 [01:09<11:53, 64.32it/s]




  8%|▊         | 4098/50000 [01:09<12:22, 61.79it/s]




  8%|▊         | 4105/50000 [01:09<12:17, 62.22it/s]




  8%|▊         | 4112/50000 [01:09<12:55, 59.20it/s]




  8%|▊         | 4120/50000 [01:09<11:55, 64.09it/s]




  8%|▊         | 4127/50000 [01:09<12:10, 62.80it/s]




  8%|▊         | 4134/50000 [01:09<13:01, 58.69it/s]




  8%|▊         | 4141/50000 [01:09<12:42, 60.13it/s]




  8%|▊         | 4148/50000 [01:10<12:30, 61.07it/s]




  8%|▊         | 4155/50000 [01:10<13:02, 58.61it/s]




  8%|▊         | 4163/50000 [01:10<12:05, 63.16it/s]




  8%|▊         | 4170/50000 [01:10<12:21, 61.84it/s]




  8%|▊         | 4177/50000 [01:10<12:18, 62.08it/s]




  8%|▊         | 4184/50000 [01:10<12:26, 61.36it/s]




  8%|▊         | 4191/50000 [01:10<12:10, 62.74it/s]




  8%|▊         | 4198/50000 [01:10<12:04, 63.23it/s]




  8%|▊         | 4206/50000 [01:10<11:25, 66.84it/s]




  8%|▊         | 4213/50000 [01:11<12:00, 63.57it/s]




  8%|▊         | 4220/50000 [01:11<12:22, 61.63it/s]




  8%|▊         | 4227/50000 [01:11<12:50, 59.42it/s]




  8%|▊         | 4234/50000 [01:11<13:13, 57.68it/s]




  8%|▊         | 4240/50000 [01:11<13:17, 57.40it/s]




  8%|▊         | 4246/50000 [01:11<13:57, 54.63it/s]




  9%|▊         | 4252/50000 [01:11<13:41, 55.71it/s]




  9%|▊         | 4258/50000 [01:11<14:48, 51.46it/s]




  9%|▊         | 4264/50000 [01:12<15:55, 47.85it/s]




  9%|▊         | 4270/50000 [01:12<15:02, 50.66it/s]




  9%|▊         | 4276/50000 [01:12<14:40, 51.94it/s]




  9%|▊         | 4282/50000 [01:12<14:31, 52.46it/s]




  9%|▊         | 4289/50000 [01:12<13:46, 55.28it/s]




  9%|▊         | 4296/50000 [01:12<13:01, 58.46it/s]




  9%|▊         | 4302/50000 [01:12<13:31, 56.31it/s]




  9%|▊         | 4309/50000 [01:12<12:55, 58.95it/s]




  9%|▊         | 4316/50000 [01:12<12:19, 61.76it/s]




  9%|▊         | 4323/50000 [01:13<12:12, 62.36it/s]




  9%|▊         | 4330/50000 [01:13<11:48, 64.43it/s]




  9%|▊         | 4337/50000 [01:13<11:39, 65.32it/s]




  9%|▊         | 4344/50000 [01:13<11:51, 64.15it/s]




  9%|▊         | 4351/50000 [01:13<11:46, 64.57it/s]




  9%|▊         | 4358/50000 [01:13<12:20, 61.63it/s]




  9%|▊         | 4365/50000 [01:13<12:46, 59.57it/s]




  9%|▊         | 4372/50000 [01:13<13:19, 57.07it/s]




  9%|▉         | 4378/50000 [01:13<13:27, 56.50it/s]




  9%|▉         | 4385/50000 [01:14<12:47, 59.40it/s]




  9%|▉         | 4392/50000 [01:14<12:28, 60.97it/s]




  9%|▉         | 4400/50000 [01:14<12:13, 62.19it/s]




  9%|▉         | 4407/50000 [01:14<12:30, 60.71it/s]




  9%|▉         | 4414/50000 [01:14<12:23, 61.29it/s]




  9%|▉         | 4421/50000 [01:14<12:24, 61.21it/s]




  9%|▉         | 4428/50000 [01:14<12:15, 61.97it/s]




  9%|▉         | 4435/50000 [01:14<12:07, 62.67it/s]




  9%|▉         | 4444/50000 [01:14<11:13, 67.60it/s]




  9%|▉         | 4451/50000 [01:15<12:24, 61.20it/s]




  9%|▉         | 4458/50000 [01:15<12:17, 61.71it/s]




  9%|▉         | 4465/50000 [01:15<12:08, 62.49it/s]




  9%|▉         | 4472/50000 [01:15<12:05, 62.71it/s]




  9%|▉         | 4479/50000 [01:15<11:56, 63.55it/s]




  9%|▉         | 4486/50000 [01:15<12:01, 63.11it/s]




  9%|▉         | 4493/50000 [01:15<12:06, 62.64it/s]




  9%|▉         | 4501/50000 [01:15<11:49, 64.15it/s]




  9%|▉         | 4508/50000 [01:16<12:01, 63.01it/s]




  9%|▉         | 4515/50000 [01:16<12:34, 60.31it/s]




  9%|▉         | 4522/50000 [01:16<12:22, 61.24it/s]




  9%|▉         | 4529/50000 [01:16<12:34, 60.23it/s]




  9%|▉         | 4536/50000 [01:16<12:55, 58.65it/s]




  9%|▉         | 4543/50000 [01:16<12:36, 60.05it/s]




  9%|▉         | 4550/50000 [01:16<12:39, 59.81it/s]




  9%|▉         | 4557/50000 [01:16<12:51, 58.88it/s]




  9%|▉         | 4564/50000 [01:16<12:46, 59.24it/s]




  9%|▉         | 4571/50000 [01:17<12:17, 61.57it/s]




  9%|▉         | 4578/50000 [01:17<12:19, 61.41it/s]




  9%|▉         | 4586/50000 [01:17<11:51, 63.85it/s]




  9%|▉         | 4593/50000 [01:17<12:14, 61.81it/s]




  9%|▉         | 4600/50000 [01:17<13:46, 54.93it/s]




  9%|▉         | 4606/50000 [01:17<13:58, 54.12it/s]




  9%|▉         | 4612/50000 [01:17<13:44, 55.03it/s]




  9%|▉         | 4618/50000 [01:17<14:42, 51.41it/s]




  9%|▉         | 4624/50000 [01:18<15:39, 48.32it/s]




  9%|▉         | 4629/50000 [01:18<17:03, 44.31it/s]




  9%|▉         | 4634/50000 [01:18<17:03, 44.33it/s]




  9%|▉         | 4639/50000 [01:18<16:45, 45.12it/s]




  9%|▉         | 4644/50000 [01:18<16:32, 45.68it/s]




  9%|▉         | 4650/50000 [01:18<15:41, 48.17it/s]




  9%|▉         | 4656/50000 [01:18<15:46, 47.93it/s]




  9%|▉         | 4662/50000 [01:18<15:26, 48.93it/s]




  9%|▉         | 4668/50000 [01:18<15:01, 50.27it/s]




  9%|▉         | 4675/50000 [01:19<14:08, 53.44it/s]




  9%|▉         | 4682/50000 [01:19<13:15, 57.00it/s]




  9%|▉         | 4689/50000 [01:19<12:35, 59.94it/s]




  9%|▉         | 4696/50000 [01:19<12:18, 61.37it/s]




  9%|▉         | 4703/50000 [01:19<12:09, 62.08it/s]




  9%|▉         | 4710/50000 [01:19<11:57, 63.10it/s]




  9%|▉         | 4718/50000 [01:19<11:13, 67.24it/s]




  9%|▉         | 4725/50000 [01:19<12:05, 62.44it/s]




  9%|▉         | 4732/50000 [01:19<12:13, 61.68it/s]




  9%|▉         | 4739/50000 [01:20<13:09, 57.34it/s]




  9%|▉         | 4745/50000 [01:20<13:22, 56.40it/s]




 10%|▉         | 4752/50000 [01:20<13:02, 57.85it/s]




 10%|▉         | 4758/50000 [01:20<14:00, 53.82it/s]




 10%|▉         | 4764/50000 [01:20<14:30, 51.97it/s]




 10%|▉         | 4770/50000 [01:20<15:15, 49.38it/s]




 10%|▉         | 4776/50000 [01:20<15:29, 48.64it/s]




 10%|▉         | 4782/50000 [01:20<14:42, 51.21it/s]




 10%|▉         | 4788/50000 [01:21<14:23, 52.35it/s]




 10%|▉         | 4796/50000 [01:21<13:12, 57.06it/s]




 10%|▉         | 4803/50000 [01:21<12:48, 58.85it/s]




 10%|▉         | 4810/50000 [01:21<12:39, 59.49it/s]




 10%|▉         | 4817/50000 [01:21<12:53, 58.44it/s]




 10%|▉         | 4824/50000 [01:21<12:33, 59.96it/s]




 10%|▉         | 4831/50000 [01:21<12:25, 60.58it/s]




 10%|▉         | 4839/50000 [01:21<11:44, 64.13it/s]




 10%|▉         | 4846/50000 [01:22<12:31, 60.09it/s]




 10%|▉         | 4853/50000 [01:22<12:52, 58.41it/s]




 10%|▉         | 4860/50000 [01:22<12:19, 61.02it/s]




 10%|▉         | 4867/50000 [01:22<12:09, 61.83it/s]




 10%|▉         | 4874/50000 [01:22<11:59, 62.75it/s]




 10%|▉         | 4882/50000 [01:22<11:39, 64.54it/s]




 10%|▉         | 4889/50000 [01:22<11:38, 64.58it/s]




 10%|▉         | 4896/50000 [01:22<11:51, 63.39it/s]




 10%|▉         | 4903/50000 [01:22<11:46, 63.86it/s]




 10%|▉         | 4910/50000 [01:23<11:55, 62.98it/s]




 10%|▉         | 4918/50000 [01:23<11:31, 65.16it/s]




 10%|▉         | 4925/50000 [01:23<12:08, 61.92it/s]




 10%|▉         | 4932/50000 [01:23<12:22, 60.69it/s]




 10%|▉         | 4939/50000 [01:23<12:07, 61.92it/s]




 10%|▉         | 4946/50000 [01:23<12:10, 61.65it/s]




 10%|▉         | 4953/50000 [01:23<11:55, 62.95it/s]




 10%|▉         | 4960/50000 [01:23<11:37, 64.60it/s]




 10%|▉         | 4967/50000 [01:23<11:57, 62.81it/s]




 10%|▉         | 4974/50000 [01:24<12:22, 60.65it/s]




 10%|▉         | 4982/50000 [01:24<11:39, 64.34it/s]




 10%|▉         | 4989/50000 [01:24<11:31, 65.07it/s]




 10%|▉         | 4996/50000 [01:24<11:49, 63.46it/s]




 10%|█         | 5003/50000 [01:24<11:38, 64.44it/s]




 10%|█         | 5010/50000 [01:24<11:34, 64.77it/s]




 10%|█         | 5017/50000 [01:24<11:27, 65.47it/s]




 10%|█         | 5024/50000 [01:24<11:33, 64.87it/s]




 10%|█         | 5031/50000 [01:24<11:34, 64.73it/s]




 10%|█         | 5038/50000 [01:25<11:29, 65.16it/s]




 10%|█         | 5045/50000 [01:25<11:28, 65.29it/s]




 10%|█         | 5052/50000 [01:25<11:27, 65.38it/s]




 10%|█         | 5059/50000 [01:25<11:30, 65.08it/s]




 10%|█         | 5066/50000 [01:25<11:53, 62.95it/s]




 10%|█         | 5073/50000 [01:25<11:33, 64.78it/s]




 10%|█         | 5080/50000 [01:25<11:25, 65.57it/s]




 10%|█         | 5088/50000 [01:25<11:04, 67.62it/s]




 10%|█         | 5095/50000 [01:25<11:19, 66.13it/s]




 10%|█         | 5102/50000 [01:25<11:32, 64.87it/s]




 10%|█         | 5109/50000 [01:26<12:06, 61.79it/s]




 10%|█         | 5116/50000 [01:26<11:58, 62.45it/s]




 10%|█         | 5123/50000 [01:26<11:42, 63.88it/s]




 10%|█         | 5130/50000 [01:26<11:30, 65.01it/s]




 10%|█         | 5137/50000 [01:26<11:43, 63.76it/s]




 10%|█         | 5144/50000 [01:26<11:28, 65.20it/s]




 10%|█         | 5151/50000 [01:26<11:34, 64.59it/s]




 10%|█         | 5159/50000 [01:26<11:05, 67.39it/s]




 10%|█         | 5166/50000 [01:26<11:20, 65.90it/s]




 10%|█         | 5173/50000 [01:27<11:57, 62.47it/s]




 10%|█         | 5180/50000 [01:27<12:14, 61.06it/s]




 10%|█         | 5190/50000 [01:27<11:03, 67.50it/s]




 10%|█         | 5197/50000 [01:27<11:01, 67.78it/s]




 10%|█         | 5204/50000 [01:27<11:25, 65.31it/s]




 10%|█         | 5213/50000 [01:27<10:33, 70.71it/s]




 10%|█         | 5221/50000 [01:27<10:55, 68.26it/s]




 10%|█         | 5230/50000 [01:27<10:31, 70.90it/s]




 10%|█         | 5238/50000 [01:28<10:46, 69.26it/s]




 10%|█         | 5246/50000 [01:28<11:01, 67.63it/s]




 11%|█         | 5253/50000 [01:28<11:15, 66.25it/s]




 11%|█         | 5262/50000 [01:28<10:30, 70.97it/s]




 11%|█         | 5270/50000 [01:28<10:46, 69.14it/s]




 11%|█         | 5278/50000 [01:28<10:38, 70.05it/s]




 11%|█         | 5286/50000 [01:28<10:29, 71.08it/s]




 11%|█         | 5294/50000 [01:28<10:27, 71.24it/s]




 11%|█         | 5302/50000 [01:28<10:24, 71.54it/s]




 11%|█         | 5310/50000 [01:29<10:54, 68.28it/s]




 11%|█         | 5317/50000 [01:29<11:29, 64.83it/s]




 11%|█         | 5324/50000 [01:29<11:24, 65.24it/s]




 11%|█         | 5331/50000 [01:29<11:35, 64.27it/s]




 11%|█         | 5338/50000 [01:29<12:01, 61.92it/s]




 11%|█         | 5346/50000 [01:29<11:23, 65.33it/s]




 11%|█         | 5353/50000 [01:29<11:37, 63.98it/s]




 11%|█         | 5360/50000 [01:29<11:59, 62.06it/s]




 11%|█         | 5367/50000 [01:29<11:36, 64.12it/s]




 11%|█         | 5374/50000 [01:30<11:38, 63.85it/s]




 11%|█         | 5381/50000 [01:30<11:46, 63.15it/s]




 11%|█         | 5388/50000 [01:30<11:53, 62.50it/s]




 11%|█         | 5396/50000 [01:30<11:18, 65.78it/s]




 11%|█         | 5404/50000 [01:30<11:08, 66.67it/s]




 11%|█         | 5411/50000 [01:30<11:18, 65.69it/s]




 11%|█         | 5418/50000 [01:30<11:43, 63.35it/s]




 11%|█         | 5426/50000 [01:30<11:12, 66.27it/s]




 11%|█         | 5433/50000 [01:30<11:10, 66.44it/s]




 11%|█         | 5440/50000 [01:31<11:59, 61.89it/s]




 11%|█         | 5449/50000 [01:31<11:04, 67.09it/s]




 11%|█         | 5457/50000 [01:31<10:50, 68.47it/s]




 11%|█         | 5464/50000 [01:31<11:41, 63.45it/s]




 11%|█         | 5471/50000 [01:31<12:33, 59.07it/s]




 11%|█         | 5478/50000 [01:31<13:57, 53.14it/s]




 11%|█         | 5484/50000 [01:31<14:06, 52.61it/s]




 11%|█         | 5491/50000 [01:31<13:17, 55.80it/s]




 11%|█         | 5497/50000 [01:32<13:09, 56.40it/s]




 11%|█         | 5503/50000 [01:32<13:18, 55.72it/s]




 11%|█         | 5509/50000 [01:32<13:09, 56.34it/s]




 11%|█         | 5516/50000 [01:32<12:24, 59.73it/s]




 11%|█         | 5523/50000 [01:32<11:58, 61.86it/s]




 11%|█         | 5530/50000 [01:32<11:48, 62.77it/s]




 11%|█         | 5538/50000 [01:32<11:23, 65.03it/s]




 11%|█         | 5545/50000 [01:32<11:19, 65.47it/s]




 11%|█         | 5552/50000 [01:32<11:53, 62.33it/s]




 11%|█         | 5559/50000 [01:33<11:30, 64.41it/s]




 11%|█         | 5567/50000 [01:33<11:12, 66.08it/s]




 11%|█         | 5576/50000 [01:33<10:24, 71.17it/s]




 11%|█         | 5585/50000 [01:33<10:00, 73.94it/s]




 11%|█         | 5593/50000 [01:33<10:03, 73.63it/s]




 11%|█         | 5601/50000 [01:33<10:39, 69.41it/s]




 11%|█         | 5609/50000 [01:33<12:21, 59.86it/s]




 11%|█         | 5616/50000 [01:33<13:57, 52.98it/s]




 11%|█         | 5622/50000 [01:34<14:46, 50.08it/s]




 11%|█▏        | 5628/50000 [01:34<14:23, 51.38it/s]




 11%|█▏        | 5634/50000 [01:34<13:58, 52.88it/s]




 11%|█▏        | 5641/50000 [01:34<13:13, 55.88it/s]




 11%|█▏        | 5648/50000 [01:34<12:31, 59.05it/s]




 11%|█▏        | 5655/50000 [01:34<12:30, 59.10it/s]




 11%|█▏        | 5662/50000 [01:34<12:21, 59.82it/s]




 11%|█▏        | 5669/50000 [01:34<13:17, 55.60it/s]




 11%|█▏        | 5676/50000 [01:35<12:38, 58.41it/s]




 11%|█▏        | 5682/50000 [01:35<13:16, 55.65it/s]




 11%|█▏        | 5689/50000 [01:35<12:51, 57.45it/s]




 11%|█▏        | 5696/50000 [01:35<12:16, 60.13it/s]




 11%|█▏        | 5703/50000 [01:35<13:25, 55.00it/s]




 11%|█▏        | 5711/50000 [01:35<12:28, 59.18it/s]




 11%|█▏        | 5718/50000 [01:35<12:02, 61.30it/s]




 11%|█▏        | 5725/50000 [01:35<12:31, 58.90it/s]




 11%|█▏        | 5732/50000 [01:35<12:16, 60.14it/s]




 11%|█▏        | 5739/50000 [01:36<15:15, 48.35it/s]




 11%|█▏        | 5745/50000 [01:36<16:29, 44.72it/s]




 12%|█▏        | 5750/50000 [01:36<19:09, 38.50it/s]




 12%|█▏        | 5755/50000 [01:36<22:21, 32.97it/s]




 12%|█▏        | 5759/50000 [01:36<22:23, 32.94it/s]




 12%|█▏        | 5763/50000 [01:36<22:46, 32.36it/s]




 12%|█▏        | 5767/50000 [01:37<21:47, 33.83it/s]




 12%|█▏        | 5771/50000 [01:37<23:21, 31.56it/s]




 12%|█▏        | 5775/50000 [01:37<23:37, 31.20it/s]




 12%|█▏        | 5781/50000 [01:37<20:30, 35.94it/s]




 12%|█▏        | 5786/50000 [01:37<19:17, 38.20it/s]




 12%|█▏        | 5791/50000 [01:37<18:21, 40.15it/s]




 12%|█▏        | 5799/50000 [01:37<15:55, 46.28it/s]




 12%|█▏        | 5806/50000 [01:37<14:40, 50.22it/s]




 12%|█▏        | 5812/50000 [01:38<16:44, 43.97it/s]




 12%|█▏        | 5817/50000 [01:38<16:54, 43.55it/s]




 12%|█▏        | 5823/50000 [01:38<15:40, 46.98it/s]




 12%|█▏        | 5830/50000 [01:38<14:18, 51.42it/s]




 12%|█▏        | 5836/50000 [01:38<13:57, 52.70it/s]




 12%|█▏        | 5843/50000 [01:38<13:00, 56.55it/s]




 12%|█▏        | 5850/50000 [01:38<12:17, 59.89it/s]




 12%|█▏        | 5857/50000 [01:38<12:50, 57.28it/s]




 12%|█▏        | 5863/50000 [01:38<13:36, 54.04it/s]




 12%|█▏        | 5869/50000 [01:39<14:02, 52.39it/s]




 12%|█▏        | 5875/50000 [01:39<13:31, 54.36it/s]




 12%|█▏        | 5883/50000 [01:39<12:18, 59.77it/s]




 12%|█▏        | 5891/50000 [01:39<11:28, 64.10it/s]




 12%|█▏        | 5899/50000 [01:39<11:37, 63.21it/s]




 12%|█▏        | 5906/50000 [01:39<12:15, 59.98it/s]




 12%|█▏        | 5913/50000 [01:39<12:54, 56.92it/s]




 12%|█▏        | 5920/50000 [01:39<12:29, 58.83it/s]




 12%|█▏        | 5928/50000 [01:40<11:50, 62.01it/s]




 12%|█▏        | 5935/50000 [01:40<13:24, 54.78it/s]




 12%|█▏        | 5941/50000 [01:40<14:43, 49.85it/s]




 12%|█▏        | 5947/50000 [01:40<19:25, 37.80it/s]




 12%|█▏        | 5952/50000 [01:40<20:37, 35.59it/s]




 12%|█▏        | 5958/50000 [01:40<18:58, 38.68it/s]




 12%|█▏        | 5963/50000 [01:41<20:55, 35.08it/s]




 12%|█▏        | 5967/50000 [01:41<22:26, 32.71it/s]




 12%|█▏        | 5971/50000 [01:41<23:31, 31.19it/s]




 12%|█▏        | 5976/50000 [01:41<21:40, 33.86it/s]




 12%|█▏        | 5980/50000 [01:41<21:37, 33.94it/s]




 12%|█▏        | 5984/50000 [01:41<20:56, 35.02it/s]




 12%|█▏        | 5988/50000 [01:41<20:22, 36.01it/s]




 12%|█▏        | 5992/50000 [01:41<20:20, 36.05it/s]




 12%|█▏        | 5996/50000 [01:41<20:19, 36.07it/s]




 12%|█▏        | 6000/50000 [01:42<23:48, 30.81it/s]




 12%|█▏        | 6006/50000 [01:42<20:45, 35.33it/s]




 12%|█▏        | 6012/50000 [01:42<18:14, 40.21it/s]




 12%|█▏        | 6018/50000 [01:42<16:30, 44.40it/s]




 12%|█▏        | 6024/50000 [01:42<15:39, 46.83it/s]




 12%|█▏        | 6030/50000 [01:42<16:14, 45.14it/s]




 12%|█▏        | 6035/50000 [01:42<17:09, 42.72it/s]




 12%|█▏        | 6042/50000 [01:42<15:17, 47.91it/s]




 12%|█▏        | 6048/50000 [01:43<14:36, 50.13it/s]




 12%|█▏        | 6056/50000 [01:43<13:02, 56.19it/s]




 12%|█▏        | 6063/50000 [01:43<13:32, 54.05it/s]




 12%|█▏        | 6070/50000 [01:43<13:00, 56.32it/s]




 12%|█▏        | 6078/50000 [01:43<11:51, 61.77it/s]




 12%|█▏        | 6085/50000 [01:43<11:31, 63.47it/s]




 12%|█▏        | 6092/50000 [01:43<11:16, 64.89it/s]




 12%|█▏        | 6100/50000 [01:43<10:42, 68.31it/s]




 12%|█▏        | 6108/50000 [01:43<10:40, 68.56it/s]




 12%|█▏        | 6115/50000 [01:44<10:54, 67.06it/s]




 12%|█▏        | 6123/50000 [01:44<10:58, 66.66it/s]




 12%|█▏        | 6131/50000 [01:44<10:46, 67.90it/s]




 12%|█▏        | 6138/50000 [01:44<12:11, 59.96it/s]




 12%|█▏        | 6145/50000 [01:44<11:41, 62.53it/s]




 12%|█▏        | 6152/50000 [01:44<11:30, 63.48it/s]




 12%|█▏        | 6160/50000 [01:44<10:55, 66.90it/s]




 12%|█▏        | 6167/50000 [01:44<10:46, 67.77it/s]




 12%|█▏        | 6174/50000 [01:44<11:30, 63.46it/s]




 12%|█▏        | 6181/50000 [01:45<12:31, 58.34it/s]




 12%|█▏        | 6188/50000 [01:45<12:31, 58.28it/s]




 12%|█▏        | 6195/50000 [01:45<12:11, 59.91it/s]




 12%|█▏        | 6203/50000 [01:45<11:18, 64.52it/s]




 12%|█▏        | 6210/50000 [01:45<11:11, 65.20it/s]




 12%|█▏        | 6218/50000 [01:45<10:43, 68.02it/s]




 12%|█▏        | 6225/50000 [01:45<11:05, 65.76it/s]




 12%|█▏        | 6232/50000 [01:45<10:54, 66.84it/s]




 12%|█▏        | 6240/50000 [01:45<10:44, 67.86it/s]




 12%|█▏        | 6247/50000 [01:46<10:58, 66.40it/s]




 13%|█▎        | 6254/50000 [01:46<11:12, 65.05it/s]




 13%|█▎        | 6263/50000 [01:46<10:29, 69.52it/s]




 13%|█▎        | 6271/50000 [01:46<10:23, 70.14it/s]




 13%|█▎        | 6279/50000 [01:46<10:35, 68.76it/s]




 13%|█▎        | 6286/50000 [01:46<10:47, 67.54it/s]




 13%|█▎        | 6293/50000 [01:46<10:51, 67.04it/s]




 13%|█▎        | 6300/50000 [01:46<11:19, 64.32it/s]




 13%|█▎        | 6308/50000 [01:46<11:05, 65.69it/s]




 13%|█▎        | 6315/50000 [01:47<11:48, 61.68it/s]




 13%|█▎        | 6322/50000 [01:47<11:27, 63.49it/s]




 13%|█▎        | 6330/50000 [01:47<11:06, 65.57it/s]




 13%|█▎        | 6338/50000 [01:47<10:44, 67.78it/s]




 13%|█▎        | 6345/50000 [01:47<10:57, 66.35it/s]




 13%|█▎        | 6352/50000 [01:47<10:54, 66.69it/s]




 13%|█▎        | 6360/50000 [01:47<11:02, 65.83it/s]




 13%|█▎        | 6368/50000 [01:47<10:41, 68.01it/s]




 13%|█▎        | 6376/50000 [01:48<10:33, 68.87it/s]




 13%|█▎        | 6383/50000 [01:48<10:39, 68.25it/s]




 13%|█▎        | 6391/50000 [01:48<10:43, 67.73it/s]




 13%|█▎        | 6399/50000 [01:48<10:41, 67.97it/s]




 13%|█▎        | 6406/50000 [01:48<11:06, 65.36it/s]




 13%|█▎        | 6413/50000 [01:48<11:28, 63.31it/s]




 13%|█▎        | 6422/50000 [01:48<10:31, 69.06it/s]




 13%|█▎        | 6431/50000 [01:48<09:55, 73.20it/s]




 13%|█▎        | 6439/50000 [01:48<10:08, 71.55it/s]




 13%|█▎        | 6447/50000 [01:49<10:11, 71.19it/s]




 13%|█▎        | 6455/50000 [01:49<10:36, 68.40it/s]




 13%|█▎        | 6463/50000 [01:49<10:23, 69.79it/s]




 13%|█▎        | 6471/50000 [01:49<10:11, 71.17it/s]




 13%|█▎        | 6479/50000 [01:49<11:12, 64.72it/s]




 13%|█▎        | 6486/50000 [01:49<12:28, 58.15it/s]




 13%|█▎        | 6493/50000 [01:49<12:45, 56.87it/s]




 13%|█▎        | 6501/50000 [01:49<12:02, 60.21it/s]




 13%|█▎        | 6508/50000 [01:50<11:45, 61.64it/s]




 13%|█▎        | 6515/50000 [01:50<11:40, 62.11it/s]




 13%|█▎        | 6522/50000 [01:50<11:24, 63.54it/s]




 13%|█▎        | 6529/50000 [01:50<11:27, 63.28it/s]




 13%|█▎        | 6537/50000 [01:50<11:14, 64.46it/s]




 13%|█▎        | 6544/50000 [01:50<11:51, 61.08it/s]




 13%|█▎        | 6551/50000 [01:50<11:42, 61.87it/s]




 13%|█▎        | 6560/50000 [01:50<10:44, 67.39it/s]




 13%|█▎        | 6568/50000 [01:50<10:23, 69.68it/s]




 13%|█▎        | 6576/50000 [01:51<10:01, 72.15it/s]




 13%|█▎        | 6584/50000 [01:51<09:53, 73.17it/s]




 13%|█▎        | 6592/50000 [01:51<09:48, 73.70it/s]




 13%|█▎        | 6601/50000 [01:51<09:25, 76.74it/s]




 13%|█▎        | 6609/50000 [01:51<09:24, 76.85it/s]




 13%|█▎        | 6617/50000 [01:51<09:40, 74.78it/s]




 13%|█▎        | 6625/50000 [01:51<09:39, 74.84it/s]




 13%|█▎        | 6634/50000 [01:51<09:13, 78.30it/s]




 13%|█▎        | 6642/50000 [01:51<09:56, 72.74it/s]




 13%|█▎        | 6650/50000 [01:52<10:15, 70.42it/s]




 13%|█▎        | 6658/50000 [01:52<10:37, 68.03it/s]




 13%|█▎        | 6665/50000 [01:52<10:49, 66.69it/s]




 13%|█▎        | 6672/50000 [01:52<11:05, 65.06it/s]




 13%|█▎        | 6679/50000 [01:52<11:15, 64.15it/s]




 13%|█▎        | 6686/50000 [01:52<11:36, 62.17it/s]




 13%|█▎        | 6693/50000 [01:52<11:42, 61.66it/s]




 13%|█▎        | 6701/50000 [01:52<11:16, 64.04it/s]




 13%|█▎        | 6708/50000 [01:52<11:16, 63.97it/s]




 13%|█▎        | 6716/50000 [01:53<10:43, 67.25it/s]




 13%|█▎        | 6725/50000 [01:53<10:08, 71.09it/s]




 13%|█▎        | 6735/50000 [01:53<09:33, 75.43it/s]




 13%|█▎        | 6743/50000 [01:53<10:02, 71.75it/s]




 14%|█▎        | 6751/50000 [01:53<10:24, 69.20it/s]




 14%|█▎        | 6760/50000 [01:53<10:04, 71.57it/s]




 14%|█▎        | 6768/50000 [01:53<09:49, 73.35it/s]




 14%|█▎        | 6776/50000 [01:53<09:45, 73.78it/s]




 14%|█▎        | 6784/50000 [01:53<10:09, 70.89it/s]




 14%|█▎        | 6792/50000 [01:54<10:00, 71.97it/s]




 14%|█▎        | 6800/50000 [01:54<10:16, 70.07it/s]




 14%|█▎        | 6808/50000 [01:54<10:31, 68.36it/s]




 14%|█▎        | 6817/50000 [01:54<10:10, 70.78it/s]




 14%|█▎        | 6825/50000 [01:54<09:59, 71.98it/s]




 14%|█▎        | 6833/50000 [01:54<10:26, 68.91it/s]




 14%|█▎        | 6842/50000 [01:54<10:00, 71.87it/s]




 14%|█▎        | 6850/50000 [01:54<09:54, 72.57it/s]




 14%|█▎        | 6858/50000 [01:54<09:53, 72.68it/s]




 14%|█▎        | 6866/50000 [01:55<09:46, 73.55it/s]




 14%|█▎        | 6874/50000 [01:55<09:33, 75.22it/s]




 14%|█▍        | 6883/50000 [01:55<09:17, 77.34it/s]




 14%|█▍        | 6892/50000 [01:55<08:56, 80.31it/s]




 14%|█▍        | 6901/50000 [01:55<08:49, 81.38it/s]




 14%|█▍        | 6910/50000 [01:55<08:49, 81.36it/s]




 14%|█▍        | 6919/50000 [01:55<09:01, 79.62it/s]




 14%|█▍        | 6927/50000 [01:55<09:15, 77.48it/s]




 14%|█▍        | 6936/50000 [01:55<09:06, 78.80it/s]




 14%|█▍        | 6944/50000 [01:56<09:45, 73.55it/s]




 14%|█▍        | 6952/50000 [01:56<09:51, 72.76it/s]




 14%|█▍        | 6961/50000 [01:56<09:21, 76.66it/s]




 14%|█▍        | 6970/50000 [01:56<09:08, 78.50it/s]




 14%|█▍        | 6978/50000 [01:56<09:09, 78.25it/s]




 14%|█▍        | 6986/50000 [01:56<09:36, 74.65it/s]




 14%|█▍        | 6994/50000 [01:56<09:25, 76.02it/s]




 14%|█▍        | 7002/50000 [01:56<09:52, 72.62it/s]




 14%|█▍        | 7011/50000 [01:56<09:38, 74.28it/s]




 14%|█▍        | 7019/50000 [01:57<10:23, 68.91it/s]




 14%|█▍        | 7029/50000 [01:57<09:39, 74.19it/s]




 14%|█▍        | 7038/50000 [01:57<09:15, 77.36it/s]




 14%|█▍        | 7048/50000 [01:57<08:47, 81.43it/s]




 14%|█▍        | 7057/50000 [01:57<09:31, 75.09it/s]




 14%|█▍        | 7066/50000 [01:57<09:17, 77.05it/s]




 14%|█▍        | 7074/50000 [01:57<09:31, 75.13it/s]




 14%|█▍        | 7082/50000 [01:57<09:23, 76.15it/s]




 14%|█▍        | 7090/50000 [01:58<09:42, 73.70it/s]




 14%|█▍        | 7098/50000 [01:58<09:29, 75.33it/s]




 14%|█▍        | 7106/50000 [01:58<09:19, 76.61it/s]




 14%|█▍        | 7115/50000 [01:58<09:11, 77.76it/s]




 14%|█▍        | 7123/50000 [01:58<09:40, 73.92it/s]




 14%|█▍        | 7131/50000 [01:58<09:34, 74.65it/s]




 14%|█▍        | 7139/50000 [01:58<09:26, 75.69it/s]




 14%|█▍        | 7147/50000 [01:58<09:23, 76.02it/s]




 14%|█▍        | 7155/50000 [01:58<09:32, 74.85it/s]




 14%|█▍        | 7163/50000 [01:58<09:33, 74.68it/s]




 14%|█▍        | 7171/50000 [01:59<09:42, 73.53it/s]




 14%|█▍        | 7179/50000 [01:59<09:31, 74.99it/s]




 14%|█▍        | 7188/50000 [01:59<09:13, 77.37it/s]




 14%|█▍        | 7196/50000 [01:59<09:20, 76.41it/s]




 14%|█▍        | 7204/50000 [01:59<09:16, 76.85it/s]




 14%|█▍        | 7213/50000 [01:59<09:07, 78.13it/s]




 14%|█▍        | 7221/50000 [01:59<09:17, 76.71it/s]




 14%|█▍        | 7229/50000 [01:59<09:21, 76.18it/s]




 14%|█▍        | 7237/50000 [01:59<09:24, 75.81it/s]




 14%|█▍        | 7245/50000 [02:00<09:21, 76.11it/s]




 15%|█▍        | 7253/50000 [02:00<09:42, 73.37it/s]




 15%|█▍        | 7261/50000 [02:00<10:10, 69.98it/s]




 15%|█▍        | 7269/50000 [02:00<10:09, 70.09it/s]




 15%|█▍        | 7277/50000 [02:00<10:40, 66.67it/s]




 15%|█▍        | 7284/50000 [02:00<10:38, 66.92it/s]




 15%|█▍        | 7292/50000 [02:00<10:22, 68.61it/s]




 15%|█▍        | 7300/50000 [02:00<10:08, 70.22it/s]




 15%|█▍        | 7309/50000 [02:00<09:30, 74.85it/s]




 15%|█▍        | 7317/50000 [02:01<09:21, 76.05it/s]




 15%|█▍        | 7325/50000 [02:01<09:31, 74.67it/s]




 15%|█▍        | 7333/50000 [02:01<09:34, 74.27it/s]




 15%|█▍        | 7341/50000 [02:01<09:30, 74.82it/s]




 15%|█▍        | 7349/50000 [02:01<09:33, 74.33it/s]




 15%|█▍        | 7357/50000 [02:01<09:41, 73.29it/s]




 15%|█▍        | 7365/50000 [02:01<09:33, 74.40it/s]




 15%|█▍        | 7373/50000 [02:01<09:23, 75.63it/s]




 15%|█▍        | 7381/50000 [02:01<09:29, 74.79it/s]




 15%|█▍        | 7389/50000 [02:02<09:35, 74.02it/s]




 15%|█▍        | 7397/50000 [02:02<09:27, 75.13it/s]




 15%|█▍        | 7406/50000 [02:02<09:04, 78.28it/s]




 15%|█▍        | 7414/50000 [02:02<09:17, 76.38it/s]




 15%|█▍        | 7422/50000 [02:02<09:30, 74.68it/s]




 15%|█▍        | 7430/50000 [02:02<09:27, 74.97it/s]




 15%|█▍        | 7438/50000 [02:02<09:29, 74.76it/s]




 15%|█▍        | 7446/50000 [02:02<09:31, 74.41it/s]




 15%|█▍        | 7454/50000 [02:02<09:59, 71.03it/s]




 15%|█▍        | 7462/50000 [02:03<09:41, 73.15it/s]




 15%|█▍        | 7470/50000 [02:03<09:38, 73.48it/s]




 15%|█▍        | 7478/50000 [02:03<09:33, 74.10it/s]




 15%|█▍        | 7487/50000 [02:03<09:24, 75.29it/s]




 15%|█▍        | 7497/50000 [02:03<08:57, 79.06it/s]




 15%|█▌        | 7505/50000 [02:03<09:14, 76.67it/s]




 15%|█▌        | 7513/50000 [02:03<09:11, 77.03it/s]




 15%|█▌        | 7521/50000 [02:03<09:22, 75.54it/s]




 15%|█▌        | 7529/50000 [02:03<09:15, 76.44it/s]




 15%|█▌        | 7538/50000 [02:03<08:58, 78.86it/s]




 15%|█▌        | 7546/50000 [02:04<09:03, 78.10it/s]




 15%|█▌        | 7554/50000 [02:04<09:21, 75.61it/s]




 15%|█▌        | 7562/50000 [02:04<09:41, 72.94it/s]




 15%|█▌        | 7571/50000 [02:04<09:18, 76.03it/s]




 15%|█▌        | 7579/50000 [02:04<09:29, 74.44it/s]




 15%|█▌        | 7589/50000 [02:04<08:49, 80.09it/s]




 15%|█▌        | 7598/50000 [02:04<08:57, 78.88it/s]




 15%|█▌        | 7607/50000 [02:04<09:02, 78.14it/s]




 15%|█▌        | 7615/50000 [02:05<09:15, 76.28it/s]




 15%|█▌        | 7623/50000 [02:05<09:12, 76.75it/s]




 15%|█▌        | 7632/50000 [02:05<08:59, 78.47it/s]




 15%|█▌        | 7642/50000 [02:05<08:34, 82.29it/s]




 15%|█▌        | 7651/50000 [02:05<08:57, 78.77it/s]




 15%|█▌        | 7659/50000 [02:05<10:35, 66.58it/s]




 15%|█▌        | 7667/50000 [02:05<10:31, 67.07it/s]




 15%|█▌        | 7674/50000 [02:05<10:24, 67.78it/s]




 15%|█▌        | 7681/50000 [02:05<10:43, 65.79it/s]




 15%|█▌        | 7688/50000 [02:06<10:43, 65.73it/s]




 15%|█▌        | 7696/50000 [02:06<10:40, 66.06it/s]




 15%|█▌        | 7703/50000 [02:06<11:04, 63.67it/s]




 15%|█▌        | 7710/50000 [02:06<11:14, 62.69it/s]




 15%|█▌        | 7717/50000 [02:06<11:29, 61.37it/s]




 15%|█▌        | 7724/50000 [02:06<11:39, 60.48it/s]




 15%|█▌        | 7731/50000 [02:06<11:42, 60.17it/s]




 15%|█▌        | 7740/50000 [02:06<10:39, 66.12it/s]




 15%|█▌        | 7748/50000 [02:06<10:16, 68.55it/s]




 16%|█▌        | 7756/50000 [02:07<10:13, 68.90it/s]




 16%|█▌        | 7764/50000 [02:07<10:01, 70.25it/s]




 16%|█▌        | 7772/50000 [02:07<09:48, 71.79it/s]




 16%|█▌        | 7782/50000 [02:07<09:19, 75.46it/s]




 16%|█▌        | 7790/50000 [02:07<09:17, 75.72it/s]




 16%|█▌        | 7798/50000 [02:07<09:46, 71.94it/s]




 16%|█▌        | 7806/50000 [02:07<09:43, 72.32it/s]




 16%|█▌        | 7814/50000 [02:07<10:16, 68.42it/s]




 16%|█▌        | 7822/50000 [02:07<09:57, 70.58it/s]




 16%|█▌        | 7830/50000 [02:08<09:37, 73.02it/s]




 16%|█▌        | 7838/50000 [02:08<10:10, 69.03it/s]




 16%|█▌        | 7846/50000 [02:08<10:13, 68.71it/s]




 16%|█▌        | 7853/50000 [02:08<11:37, 60.40it/s]




 16%|█▌        | 7860/50000 [02:08<11:12, 62.70it/s]




 16%|█▌        | 7867/50000 [02:08<10:54, 64.42it/s]




 16%|█▌        | 7875/50000 [02:08<10:37, 66.09it/s]




 16%|█▌        | 7883/50000 [02:08<10:16, 68.35it/s]




 16%|█▌        | 7891/50000 [02:09<09:51, 71.14it/s]




 16%|█▌        | 7900/50000 [02:09<09:29, 73.91it/s]




 16%|█▌        | 7908/50000 [02:09<09:23, 74.74it/s]




 16%|█▌        | 7916/50000 [02:09<09:24, 74.49it/s]




 16%|█▌        | 7925/50000 [02:09<09:07, 76.90it/s]




 16%|█▌        | 7933/50000 [02:09<09:24, 74.50it/s]




 16%|█▌        | 7942/50000 [02:09<08:57, 78.21it/s]




 16%|█▌        | 7951/50000 [02:09<08:50, 79.32it/s]




 16%|█▌        | 7959/50000 [02:09<09:22, 74.69it/s]




 16%|█▌        | 7967/50000 [02:10<10:10, 68.81it/s]




 16%|█▌        | 7975/50000 [02:10<09:50, 71.11it/s]




 16%|█▌        | 7984/50000 [02:10<09:16, 75.56it/s]




 16%|█▌        | 7993/50000 [02:10<09:04, 77.19it/s]




 16%|█▌        | 8001/50000 [02:10<09:30, 73.66it/s]




 16%|█▌        | 8010/50000 [02:10<09:01, 77.54it/s]




 16%|█▌        | 8018/50000 [02:10<09:07, 76.75it/s]




 16%|█▌        | 8026/50000 [02:10<09:09, 76.42it/s]




 16%|█▌        | 8034/50000 [02:10<09:10, 76.20it/s]




 16%|█▌        | 8043/50000 [02:11<09:02, 77.28it/s]




 16%|█▌        | 8051/50000 [02:11<09:26, 74.03it/s]




 16%|█▌        | 8059/50000 [02:11<09:21, 74.72it/s]




 16%|█▌        | 8068/50000 [02:11<09:10, 76.19it/s]




 16%|█▌        | 8077/50000 [02:11<08:47, 79.50it/s]




 16%|█▌        | 8086/50000 [02:11<08:45, 79.82it/s]




 16%|█▌        | 8095/50000 [02:11<08:54, 78.38it/s]




 16%|█▌        | 8103/50000 [02:11<08:54, 78.46it/s]




 16%|█▌        | 8111/50000 [02:11<09:12, 75.84it/s]




 16%|█▌        | 8119/50000 [02:12<10:26, 66.88it/s]




 16%|█▋        | 8126/50000 [02:12<10:33, 66.11it/s]




 16%|█▋        | 8133/50000 [02:12<11:10, 62.44it/s]




 16%|█▋        | 8140/50000 [02:12<11:32, 60.48it/s]




 16%|█▋        | 8147/50000 [02:12<11:20, 61.52it/s]




 16%|█▋        | 8155/50000 [02:12<10:45, 64.78it/s]




 16%|█▋        | 8163/50000 [02:12<10:19, 67.53it/s]




 16%|█▋        | 8171/50000 [02:12<09:56, 70.15it/s]




 16%|█▋        | 8179/50000 [02:12<09:58, 69.85it/s]




 16%|█▋        | 8187/50000 [02:13<09:57, 70.00it/s]




 16%|█▋        | 8195/50000 [02:13<10:57, 63.61it/s]




 16%|█▋        | 8203/50000 [02:13<10:21, 67.31it/s]




 16%|█▋        | 8212/50000 [02:13<09:55, 70.14it/s]




 16%|█▋        | 8220/50000 [02:13<10:12, 68.24it/s]




 16%|█▋        | 8228/50000 [02:13<09:55, 70.13it/s]




 16%|█▋        | 8237/50000 [02:13<09:18, 74.80it/s]




 16%|█▋        | 8247/50000 [02:13<08:46, 79.24it/s]




 17%|█▋        | 8256/50000 [02:13<08:45, 79.43it/s]




 17%|█▋        | 8265/50000 [02:14<08:39, 80.30it/s]




 17%|█▋        | 8274/50000 [02:14<08:46, 79.21it/s]




 17%|█▋        | 8283/50000 [02:14<08:41, 80.04it/s]




 17%|█▋        | 8292/50000 [02:14<08:24, 82.62it/s]




 17%|█▋        | 8301/50000 [02:14<09:05, 76.37it/s]




 17%|█▋        | 8309/50000 [02:14<10:12, 68.02it/s]




 17%|█▋        | 8317/50000 [02:14<10:20, 67.16it/s]




 17%|█▋        | 8324/50000 [02:14<10:53, 63.77it/s]




 17%|█▋        | 8332/50000 [02:15<10:19, 67.27it/s]




 17%|█▋        | 8340/50000 [02:15<10:06, 68.69it/s]




 17%|█▋        | 8347/50000 [02:15<10:06, 68.73it/s]




 17%|█▋        | 8356/50000 [02:15<09:36, 72.24it/s]




 17%|█▋        | 8365/50000 [02:15<09:06, 76.12it/s]




 17%|█▋        | 8373/50000 [02:15<09:19, 74.38it/s]




 17%|█▋        | 8381/50000 [02:15<09:26, 73.49it/s]




 17%|█▋        | 8389/50000 [02:15<09:45, 71.07it/s]




 17%|█▋        | 8397/50000 [02:15<09:42, 71.43it/s]




 17%|█▋        | 8406/50000 [02:16<09:12, 75.24it/s]




 17%|█▋        | 8414/50000 [02:16<09:19, 74.32it/s]




 17%|█▋        | 8423/50000 [02:16<08:55, 77.66it/s]




 17%|█▋        | 8432/50000 [02:16<08:42, 79.55it/s]




 17%|█▋        | 8441/50000 [02:16<11:05, 62.45it/s]




 17%|█▋        | 8448/50000 [02:16<11:29, 60.26it/s]




 17%|█▋        | 8455/50000 [02:16<11:41, 59.26it/s]




 17%|█▋        | 8462/50000 [02:16<11:41, 59.24it/s]




 17%|█▋        | 8469/50000 [02:17<12:12, 56.72it/s]




 17%|█▋        | 8475/50000 [02:17<12:26, 55.60it/s]




 17%|█▋        | 8481/50000 [02:17<12:12, 56.65it/s]




 17%|█▋        | 8487/50000 [02:17<12:27, 55.50it/s]




 17%|█▋        | 8493/50000 [02:17<12:44, 54.28it/s]




 17%|█▋        | 8499/50000 [02:17<13:32, 51.08it/s]




 17%|█▋        | 8506/50000 [02:17<12:31, 55.23it/s]




 17%|█▋        | 8512/50000 [02:17<12:47, 54.03it/s]




 17%|█▋        | 8519/50000 [02:17<12:31, 55.16it/s]




 17%|█▋        | 8525/50000 [02:18<12:27, 55.48it/s]




 17%|█▋        | 8531/50000 [02:18<12:45, 54.19it/s]




 17%|█▋        | 8537/50000 [02:18<12:28, 55.39it/s]




 17%|█▋        | 8543/50000 [02:18<13:01, 53.06it/s]




 17%|█▋        | 8549/50000 [02:18<12:50, 53.83it/s]




 17%|█▋        | 8555/50000 [02:18<12:38, 54.67it/s]




 17%|█▋        | 8562/50000 [02:18<12:19, 56.03it/s]




 17%|█▋        | 8569/50000 [02:18<11:49, 58.39it/s]




 17%|█▋        | 8575/50000 [02:19<12:27, 55.41it/s]




 17%|█▋        | 8581/50000 [02:19<12:33, 54.99it/s]




 17%|█▋        | 8588/50000 [02:19<12:15, 56.33it/s]




 17%|█▋        | 8594/50000 [02:19<12:17, 56.11it/s]




 17%|█▋        | 8600/50000 [02:19<12:38, 54.61it/s]




 17%|█▋        | 8608/50000 [02:19<11:33, 59.72it/s]




 17%|█▋        | 8615/50000 [02:19<11:23, 60.58it/s]




 17%|█▋        | 8623/50000 [02:19<10:42, 64.43it/s]




 17%|█▋        | 8630/50000 [02:19<10:49, 63.72it/s]




 17%|█▋        | 8637/50000 [02:19<10:36, 64.98it/s]




 17%|█▋        | 8644/50000 [02:20<10:29, 65.71it/s]




 17%|█▋        | 8652/50000 [02:20<10:08, 67.90it/s]




 17%|█▋        | 8659/50000 [02:20<10:09, 67.77it/s]




 17%|█▋        | 8666/50000 [02:20<10:14, 67.30it/s]




 17%|█▋        | 8674/50000 [02:20<09:50, 69.93it/s]




 17%|█▋        | 8682/50000 [02:20<10:07, 67.98it/s]




 17%|█▋        | 8689/50000 [02:20<10:16, 67.05it/s]




 17%|█▋        | 8696/50000 [02:20<10:46, 63.85it/s]




 17%|█▋        | 8703/50000 [02:21<12:07, 56.78it/s]




 17%|█▋        | 8709/50000 [02:21<12:57, 53.09it/s]




 17%|█▋        | 8716/50000 [02:21<12:15, 56.11it/s]




 17%|█▋        | 8722/50000 [02:21<12:04, 56.94it/s]




 17%|█▋        | 8728/50000 [02:21<12:01, 57.21it/s]




 17%|█▋        | 8734/50000 [02:21<13:13, 52.04it/s]




 17%|█▋        | 8740/50000 [02:21<12:57, 53.08it/s]




 17%|█▋        | 8746/50000 [02:21<13:41, 50.20it/s]




 18%|█▊        | 8752/50000 [02:21<13:58, 49.20it/s]




 18%|█▊        | 8758/50000 [02:22<16:03, 42.82it/s]




 18%|█▊        | 8763/50000 [02:22<15:45, 43.61it/s]




 18%|█▊        | 8770/50000 [02:22<14:27, 47.54it/s]




 18%|█▊        | 8775/50000 [02:22<14:26, 47.60it/s]




 18%|█▊        | 8781/50000 [02:22<13:52, 49.53it/s]




 18%|█▊        | 8787/50000 [02:22<13:26, 51.10it/s]




 18%|█▊        | 8793/50000 [02:22<13:00, 52.81it/s]




 18%|█▊        | 8799/50000 [02:22<14:03, 48.82it/s]




 18%|█▊        | 8804/50000 [02:23<14:09, 48.49it/s]




 18%|█▊        | 8809/50000 [02:23<14:28, 47.44it/s]




 18%|█▊        | 8817/50000 [02:23<12:47, 53.64it/s]




 18%|█▊        | 8823/50000 [02:23<12:26, 55.14it/s]




 18%|█▊        | 8830/50000 [02:23<11:59, 57.20it/s]




 18%|█▊        | 8838/50000 [02:23<11:12, 61.17it/s]




 18%|█▊        | 8845/50000 [02:23<10:50, 63.28it/s]




 18%|█▊        | 8853/50000 [02:23<10:30, 65.25it/s]




 18%|█▊        | 8860/50000 [02:23<10:53, 62.96it/s]




 18%|█▊        | 8867/50000 [02:24<10:43, 63.90it/s]




 18%|█▊        | 8875/50000 [02:24<10:37, 64.52it/s]




 18%|█▊        | 8882/50000 [02:24<11:24, 60.09it/s]




 18%|█▊        | 8889/50000 [02:24<11:07, 61.56it/s]




 18%|█▊        | 8896/50000 [02:24<11:46, 58.17it/s]




 18%|█▊        | 8902/50000 [02:24<12:14, 55.97it/s]




 18%|█▊        | 8908/50000 [02:24<13:00, 52.66it/s]




 18%|█▊        | 8914/50000 [02:24<13:20, 51.34it/s]




 18%|█▊        | 8920/50000 [02:25<14:04, 48.63it/s]




 18%|█▊        | 8925/50000 [02:25<13:59, 48.93it/s]




 18%|█▊        | 8932/50000 [02:25<13:11, 51.91it/s]




 18%|█▊        | 8938/50000 [02:25<12:54, 52.99it/s]




 18%|█▊        | 8945/50000 [02:25<12:18, 55.56it/s]




 18%|█▊        | 8951/50000 [02:25<12:09, 56.30it/s]




 18%|█▊        | 8960/50000 [02:25<10:57, 62.38it/s]




 18%|█▊        | 8967/50000 [02:25<10:40, 64.09it/s]




 18%|█▊        | 8975/50000 [02:25<10:16, 66.59it/s]




 18%|█▊        | 8982/50000 [02:26<10:08, 67.43it/s]




 18%|█▊        | 8990/50000 [02:26<09:51, 69.35it/s]




 18%|█▊        | 8998/50000 [02:26<09:37, 70.94it/s]




 18%|█▊        | 9006/50000 [02:26<10:03, 67.89it/s]




 18%|█▊        | 9013/50000 [02:26<10:48, 63.19it/s]




 18%|█▊        | 9020/50000 [02:26<10:44, 63.55it/s]




 18%|█▊        | 9027/50000 [02:26<10:40, 63.97it/s]




 18%|█▊        | 9034/50000 [02:26<10:25, 65.54it/s]




 18%|█▊        | 9042/50000 [02:26<10:03, 67.83it/s]




 18%|█▊        | 9049/50000 [02:27<10:34, 64.55it/s]




 18%|█▊        | 9056/50000 [02:27<10:52, 62.77it/s]




 18%|█▊        | 9063/50000 [02:27<10:52, 62.74it/s]




 18%|█▊        | 9070/50000 [02:27<10:49, 62.99it/s]




 18%|█▊        | 9077/50000 [02:27<11:03, 61.65it/s]




 18%|█▊        | 9084/50000 [02:27<11:06, 61.38it/s]




 18%|█▊        | 9091/50000 [02:27<11:28, 59.41it/s]




 18%|█▊        | 9099/50000 [02:27<10:42, 63.65it/s]




 18%|█▊        | 9106/50000 [02:27<11:06, 61.36it/s]




 18%|█▊        | 9113/50000 [02:28<10:58, 62.07it/s]




 18%|█▊        | 9120/50000 [02:28<11:19, 60.17it/s]




 18%|█▊        | 9127/50000 [02:28<11:04, 61.54it/s]




 18%|█▊        | 9134/50000 [02:28<10:43, 63.55it/s]




 18%|█▊        | 9141/50000 [02:28<10:43, 63.46it/s]




 18%|█▊        | 9148/50000 [02:28<10:42, 63.56it/s]




 18%|█▊        | 9156/50000 [02:28<10:05, 67.44it/s]




 18%|█▊        | 9165/50000 [02:28<09:29, 71.76it/s]




 18%|█▊        | 9173/50000 [02:28<10:08, 67.04it/s]




 18%|█▊        | 9180/50000 [02:29<10:51, 62.67it/s]




 18%|█▊        | 9188/50000 [02:29<10:19, 65.92it/s]




 18%|█▊        | 9195/50000 [02:29<10:14, 66.38it/s]




 18%|█▊        | 9202/50000 [02:29<10:22, 65.59it/s]




 18%|█▊        | 9209/50000 [02:29<10:11, 66.71it/s]




 18%|█▊        | 9216/50000 [02:29<11:01, 61.65it/s]




 18%|█▊        | 9223/50000 [02:29<11:20, 59.89it/s]




 18%|█▊        | 9230/50000 [02:29<11:47, 57.63it/s]




 18%|█▊        | 9237/50000 [02:30<11:37, 58.46it/s]




 18%|█▊        | 9243/50000 [02:30<12:07, 56.00it/s]




 18%|█▊        | 9249/50000 [02:30<12:09, 55.89it/s]




 19%|█▊        | 9255/50000 [02:30<11:54, 57.02it/s]




 19%|█▊        | 9261/50000 [02:30<11:58, 56.70it/s]




 19%|█▊        | 9269/50000 [02:30<11:08, 60.89it/s]




 19%|█▊        | 9276/50000 [02:30<10:47, 62.90it/s]




 19%|█▊        | 9283/50000 [02:30<10:32, 64.39it/s]




 19%|█▊        | 9290/50000 [02:30<10:18, 65.84it/s]




 19%|█▊        | 9297/50000 [02:31<10:41, 63.45it/s]




 19%|█▊        | 9304/50000 [02:31<10:43, 63.21it/s]




 19%|█▊        | 9312/50000 [02:31<10:14, 66.17it/s]




 19%|█▊        | 9320/50000 [02:31<10:01, 67.62it/s]




 19%|█▊        | 9328/50000 [02:31<10:07, 66.97it/s]




 19%|█▊        | 9335/50000 [02:31<10:09, 66.74it/s]




 19%|█▊        | 9343/50000 [02:31<09:46, 69.37it/s]




 19%|█▊        | 9350/50000 [02:31<10:01, 67.60it/s]




 19%|█▊        | 9358/50000 [02:31<09:46, 69.29it/s]




 19%|█▊        | 9366/50000 [02:31<09:36, 70.53it/s]




 19%|█▊        | 9374/50000 [02:32<09:52, 68.58it/s]




 19%|█▉        | 9381/50000 [02:32<10:42, 63.26it/s]




 19%|█▉        | 9388/50000 [02:32<10:36, 63.77it/s]




 19%|█▉        | 9396/50000 [02:32<10:14, 66.10it/s]




 19%|█▉        | 9403/50000 [02:32<10:19, 65.58it/s]




 19%|█▉        | 9411/50000 [02:32<10:06, 66.95it/s]




 19%|█▉        | 9418/50000 [02:32<10:08, 66.72it/s]




 19%|█▉        | 9425/50000 [02:32<10:14, 66.01it/s]




 19%|█▉        | 9432/50000 [02:33<10:14, 66.00it/s]




 19%|█▉        | 9439/50000 [02:33<10:32, 64.16it/s]




 19%|█▉        | 9446/50000 [02:33<11:13, 60.20it/s]




 19%|█▉        | 9453/50000 [02:33<11:05, 60.92it/s]




 19%|█▉        | 9460/50000 [02:33<11:25, 59.10it/s]




 19%|█▉        | 9466/50000 [02:33<11:38, 58.04it/s]




 19%|█▉        | 9474/50000 [02:33<10:57, 61.61it/s]




 19%|█▉        | 9481/50000 [02:33<10:57, 61.60it/s]




 19%|█▉        | 9488/50000 [02:33<10:35, 63.77it/s]




 19%|█▉        | 9496/50000 [02:34<10:20, 65.29it/s]




 19%|█▉        | 9503/50000 [02:34<10:33, 63.95it/s]




 19%|█▉        | 9511/50000 [02:34<10:05, 66.90it/s]




 19%|█▉        | 9518/50000 [02:34<10:12, 66.13it/s]




 19%|█▉        | 9525/50000 [02:34<10:39, 63.29it/s]




 19%|█▉        | 9532/50000 [02:34<10:32, 63.97it/s]




 19%|█▉        | 9539/50000 [02:34<10:21, 65.07it/s]




 19%|█▉        | 9546/50000 [02:34<10:50, 62.23it/s]




 19%|█▉        | 9553/50000 [02:34<10:50, 62.19it/s]




 19%|█▉        | 9560/50000 [02:35<11:02, 61.03it/s]




 19%|█▉        | 9567/50000 [02:35<11:19, 59.53it/s]




 19%|█▉        | 9574/50000 [02:35<11:04, 60.82it/s]




 19%|█▉        | 9581/50000 [02:35<10:50, 62.09it/s]




 19%|█▉        | 9588/50000 [02:35<11:05, 60.73it/s]




 19%|█▉        | 9595/50000 [02:35<11:21, 59.28it/s]




 19%|█▉        | 9601/50000 [02:35<11:44, 57.33it/s]




 19%|█▉        | 9609/50000 [02:35<10:56, 61.54it/s]




 19%|█▉        | 9616/50000 [02:35<10:47, 62.37it/s]




 19%|█▉        | 9623/50000 [02:36<11:10, 60.21it/s]




 19%|█▉        | 9630/50000 [02:36<10:45, 62.55it/s]




 19%|█▉        | 9637/50000 [02:36<10:38, 63.26it/s]




 19%|█▉        | 9644/50000 [02:36<10:19, 65.10it/s]




 19%|█▉        | 9651/50000 [02:36<10:15, 65.52it/s]




 19%|█▉        | 9658/50000 [02:36<10:17, 65.36it/s]




 19%|█▉        | 9665/50000 [02:36<10:45, 62.46it/s]




 19%|█▉        | 9672/50000 [02:36<10:46, 62.43it/s]




 19%|█▉        | 9679/50000 [02:36<10:58, 61.21it/s]




 19%|█▉        | 9686/50000 [02:37<10:42, 62.75it/s]




 19%|█▉        | 9693/50000 [02:37<10:26, 64.36it/s]




 19%|█▉        | 9700/50000 [02:37<10:21, 64.82it/s]




 19%|█▉        | 9708/50000 [02:37<09:50, 68.25it/s]




 19%|█▉        | 9715/50000 [02:37<11:02, 60.78it/s]




 19%|█▉        | 9722/50000 [02:37<11:38, 57.70it/s]




 19%|█▉        | 9728/50000 [02:37<12:31, 53.58it/s]




 19%|█▉        | 9734/50000 [02:37<12:20, 54.35it/s]




 19%|█▉        | 9740/50000 [02:38<12:47, 52.46it/s]




 19%|█▉        | 9746/50000 [02:38<13:38, 49.20it/s]




 20%|█▉        | 9752/50000 [02:38<13:33, 49.48it/s]




 20%|█▉        | 9758/50000 [02:38<12:54, 51.99it/s]




 20%|█▉        | 9764/50000 [02:38<12:32, 53.47it/s]




 20%|█▉        | 9770/50000 [02:38<12:27, 53.83it/s]




 20%|█▉        | 9776/50000 [02:38<13:00, 51.52it/s]




 20%|█▉        | 9782/50000 [02:38<13:52, 48.32it/s]




 20%|█▉        | 9787/50000 [02:38<13:57, 48.01it/s]




 20%|█▉        | 9793/50000 [02:39<13:20, 50.21it/s]




 20%|█▉        | 9800/50000 [02:39<12:46, 52.44it/s]




 20%|█▉        | 9807/50000 [02:39<12:05, 55.40it/s]




 20%|█▉        | 9814/50000 [02:39<11:28, 58.40it/s]




 20%|█▉        | 9821/50000 [02:39<11:13, 59.62it/s]




 20%|█▉        | 9828/50000 [02:39<11:31, 58.10it/s]




 20%|█▉        | 9836/50000 [02:39<10:47, 62.01it/s]




 20%|█▉        | 9843/50000 [02:39<10:47, 62.04it/s]




 20%|█▉        | 9851/50000 [02:39<10:07, 66.07it/s]




 20%|█▉        | 9858/50000 [02:40<12:06, 55.24it/s]




 20%|█▉        | 9864/50000 [02:40<12:42, 52.65it/s]




 20%|█▉        | 9870/50000 [02:40<14:11, 47.11it/s]




 20%|█▉        | 9876/50000 [02:40<14:06, 47.41it/s]




 20%|█▉        | 9882/50000 [02:40<13:52, 48.20it/s]




 20%|█▉        | 9887/50000 [02:40<14:06, 47.38it/s]




 20%|█▉        | 9892/50000 [02:40<14:06, 47.36it/s]




 20%|█▉        | 9899/50000 [02:41<12:52, 51.89it/s]




 20%|█▉        | 9905/50000 [02:41<12:30, 53.40it/s]




 20%|█▉        | 9913/50000 [02:41<11:19, 59.03it/s]




 20%|█▉        | 9920/50000 [02:41<11:02, 60.54it/s]




 20%|█▉        | 9927/50000 [02:41<11:05, 60.22it/s]




 20%|█▉        | 9934/50000 [02:41<10:52, 61.41it/s]




 20%|█▉        | 9942/50000 [02:41<10:13, 65.25it/s]




 20%|█▉        | 9950/50000 [02:41<09:41, 68.85it/s]




 20%|█▉        | 9959/50000 [02:41<09:15, 72.08it/s]




 20%|█▉        | 9967/50000 [02:41<09:21, 71.26it/s]




 20%|█▉        | 9975/50000 [02:42<09:32, 69.88it/s]




 20%|█▉        | 9983/50000 [02:42<10:13, 65.24it/s]




 20%|█▉        | 9990/50000 [02:42<10:13, 65.16it/s]




 20%|█▉        | 9997/50000 [02:42<10:07, 65.84it/s]




 20%|██        | 10004/50000 [02:42<10:37, 62.77it/s]




 20%|██        | 10012/50000 [02:42<10:07, 65.83it/s]




 20%|██        | 10019/50000 [02:42<10:07, 65.85it/s]




 20%|██        | 10026/50000 [02:42<10:02, 66.33it/s]




 20%|██        | 10033/50000 [02:43<10:09, 65.55it/s]




 20%|██        | 10040/50000 [02:43<10:06, 65.94it/s]




 20%|██        | 10047/50000 [02:43<10:14, 64.99it/s]




 20%|██        | 10054/50000 [02:43<10:08, 65.68it/s]




 20%|██        | 10061/50000 [02:43<11:03, 60.17it/s]




 20%|██        | 10068/50000 [02:43<12:36, 52.79it/s]




 20%|██        | 10074/50000 [02:43<13:02, 51.03it/s]




 20%|██        | 10080/50000 [02:43<13:22, 49.75it/s]




 20%|██        | 10088/50000 [02:44<12:10, 54.66it/s]




 20%|██        | 10094/50000 [02:44<12:23, 53.65it/s]




 20%|██        | 10101/50000 [02:44<12:00, 55.40it/s]




 20%|██        | 10107/50000 [02:44<11:44, 56.59it/s]




 20%|██        | 10115/50000 [02:44<10:57, 60.67it/s]




 20%|██        | 10122/50000 [02:44<10:41, 62.14it/s]




 20%|██        | 10129/50000 [02:44<10:29, 63.31it/s]




 20%|██        | 10136/50000 [02:44<10:41, 62.11it/s]




 20%|██        | 10144/50000 [02:44<10:16, 64.69it/s]




 20%|██        | 10151/50000 [02:45<10:30, 63.23it/s]




 20%|██        | 10158/50000 [02:45<10:19, 64.28it/s]




 20%|██        | 10166/50000 [02:45<09:44, 68.18it/s]




 20%|██        | 10173/50000 [02:45<09:45, 67.97it/s]




 20%|██        | 10181/50000 [02:45<09:32, 69.56it/s]




 20%|██        | 10189/50000 [02:45<09:43, 68.20it/s]




 20%|██        | 10196/50000 [02:45<10:12, 64.96it/s]




 20%|██        | 10203/50000 [02:45<10:10, 65.15it/s]




 20%|██        | 10210/50000 [02:45<10:11, 65.10it/s]




 20%|██        | 10218/50000 [02:45<09:40, 68.47it/s]




 20%|██        | 10225/50000 [02:46<10:13, 64.87it/s]




 20%|██        | 10232/50000 [02:46<10:43, 61.82it/s]




 20%|██        | 10239/50000 [02:46<11:02, 60.00it/s]




 20%|██        | 10246/50000 [02:46<10:42, 61.90it/s]




 21%|██        | 10253/50000 [02:46<11:36, 57.05it/s]




 21%|██        | 10259/50000 [02:46<11:40, 56.72it/s]




 21%|██        | 10265/50000 [02:46<12:02, 55.02it/s]




 21%|██        | 10272/50000 [02:46<11:27, 57.81it/s]




 21%|██        | 10278/50000 [02:47<11:44, 56.36it/s]




 21%|██        | 10285/50000 [02:47<11:33, 57.26it/s]




 21%|██        | 10291/50000 [02:47<11:29, 57.60it/s]




 21%|██        | 10298/50000 [02:47<10:53, 60.72it/s]




 21%|██        | 10306/50000 [02:47<10:14, 64.55it/s]




 21%|██        | 10313/50000 [02:47<10:29, 63.03it/s]




 21%|██        | 10320/50000 [02:47<10:16, 64.39it/s]




 21%|██        | 10327/50000 [02:47<10:09, 65.11it/s]




 21%|██        | 10334/50000 [02:47<10:04, 65.62it/s]




 21%|██        | 10343/50000 [02:48<09:18, 70.96it/s]




 21%|██        | 10351/50000 [02:48<09:50, 67.10it/s]




 21%|██        | 10358/50000 [02:48<09:48, 67.36it/s]




 21%|██        | 10365/50000 [02:48<10:18, 64.08it/s]




 21%|██        | 10372/50000 [02:48<10:17, 64.17it/s]




 21%|██        | 10379/50000 [02:48<11:17, 58.46it/s]




 21%|██        | 10385/50000 [02:48<11:49, 55.84it/s]




 21%|██        | 10391/50000 [02:48<11:41, 56.43it/s]




 21%|██        | 10397/50000 [02:48<11:34, 57.01it/s]




 21%|██        | 10403/50000 [02:49<11:45, 56.13it/s]




 21%|██        | 10410/50000 [02:49<11:14, 58.67it/s]




 21%|██        | 10417/50000 [02:49<10:55, 60.42it/s]




 21%|██        | 10424/50000 [02:49<10:51, 60.76it/s]




 21%|██        | 10431/50000 [02:49<10:30, 62.80it/s]




 21%|██        | 10439/50000 [02:49<10:03, 65.53it/s]




 21%|██        | 10446/50000 [02:49<10:27, 63.08it/s]




 21%|██        | 10453/50000 [02:49<10:15, 64.24it/s]




 21%|██        | 10460/50000 [02:49<10:21, 63.58it/s]




 21%|██        | 10467/50000 [02:50<10:14, 64.35it/s]




 21%|██        | 10475/50000 [02:50<09:42, 67.89it/s]




 21%|██        | 10482/50000 [02:50<09:58, 66.04it/s]




 21%|██        | 10490/50000 [02:50<09:35, 68.67it/s]




 21%|██        | 10497/50000 [02:50<10:05, 65.26it/s]




 21%|██        | 10504/50000 [02:50<10:11, 64.64it/s]




 21%|██        | 10511/50000 [02:50<10:50, 60.71it/s]




 21%|██        | 10518/50000 [02:50<11:17, 58.30it/s]




 21%|██        | 10526/50000 [02:50<10:26, 62.98it/s]




 21%|██        | 10533/50000 [02:51<10:30, 62.63it/s]




 21%|██        | 10540/50000 [02:51<10:21, 63.49it/s]




 21%|██        | 10547/50000 [02:51<10:18, 63.76it/s]




 21%|██        | 10554/50000 [02:51<10:23, 63.26it/s]




 21%|██        | 10562/50000 [02:51<09:57, 66.04it/s]




 21%|██        | 10570/50000 [02:51<09:37, 68.22it/s]




 21%|██        | 10578/50000 [02:51<09:25, 69.75it/s]




 21%|██        | 10586/50000 [02:51<09:45, 67.29it/s]




 21%|██        | 10593/50000 [02:51<10:10, 64.56it/s]




 21%|██        | 10600/50000 [02:52<10:32, 62.28it/s]




 21%|██        | 10607/50000 [02:52<10:27, 62.73it/s]




 21%|██        | 10614/50000 [02:52<10:21, 63.39it/s]




 21%|██        | 10621/50000 [02:52<10:20, 63.51it/s]




 21%|██▏       | 10628/50000 [02:52<10:30, 62.41it/s]




 21%|██▏       | 10635/50000 [02:52<10:37, 61.75it/s]




 21%|██▏       | 10642/50000 [02:52<10:16, 63.88it/s]




 21%|██▏       | 10649/50000 [02:52<10:04, 65.10it/s]




 21%|██▏       | 10656/50000 [02:52<10:21, 63.31it/s]




 21%|██▏       | 10663/50000 [02:53<10:11, 64.29it/s]




 21%|██▏       | 10670/50000 [02:53<10:12, 64.24it/s]




 21%|██▏       | 10677/50000 [02:53<10:25, 62.91it/s]




 21%|██▏       | 10685/50000 [02:53<09:59, 65.61it/s]




 21%|██▏       | 10692/50000 [02:53<11:28, 57.12it/s]




 21%|██▏       | 10698/50000 [02:53<11:23, 57.50it/s]




 21%|██▏       | 10704/50000 [02:53<11:57, 54.77it/s]




 21%|██▏       | 10710/50000 [02:53<13:21, 48.99it/s]




 21%|██▏       | 10716/50000 [02:54<13:50, 47.29it/s]




 21%|██▏       | 10721/50000 [02:54<14:04, 46.50it/s]




 21%|██▏       | 10726/50000 [02:54<13:48, 47.40it/s]




 21%|██▏       | 10732/50000 [02:54<13:27, 48.66it/s]




 21%|██▏       | 10737/50000 [02:54<13:43, 47.67it/s]




 21%|██▏       | 10742/50000 [02:54<13:43, 47.70it/s]




 21%|██▏       | 10747/50000 [02:54<14:15, 45.88it/s]




 22%|██▏       | 10752/50000 [02:54<14:59, 43.63it/s]




 22%|██▏       | 10757/50000 [02:54<14:29, 45.15it/s]




 22%|██▏       | 10763/50000 [02:55<13:39, 47.87it/s]




 22%|██▏       | 10769/50000 [02:55<12:51, 50.87it/s]




 22%|██▏       | 10775/50000 [02:55<12:21, 52.92it/s]




 22%|██▏       | 10782/50000 [02:55<11:39, 56.04it/s]




 22%|██▏       | 10789/50000 [02:55<11:09, 58.53it/s]




 22%|██▏       | 10796/50000 [02:55<10:48, 60.48it/s]




 22%|██▏       | 10803/50000 [02:55<11:03, 59.11it/s]




 22%|██▏       | 10810/50000 [02:55<10:38, 61.40it/s]




 22%|██▏       | 10817/50000 [02:55<10:46, 60.65it/s]




 22%|██▏       | 10824/50000 [02:56<10:59, 59.37it/s]




 22%|██▏       | 10830/50000 [02:56<11:35, 56.34it/s]




 22%|██▏       | 10836/50000 [02:56<12:29, 52.27it/s]




 22%|██▏       | 10842/50000 [02:56<14:04, 46.36it/s]




 22%|██▏       | 10847/50000 [02:56<16:38, 39.20it/s]




 22%|██▏       | 10852/50000 [02:56<16:50, 38.73it/s]




 22%|██▏       | 10857/50000 [02:56<16:09, 40.36it/s]




 22%|██▏       | 10863/50000 [02:57<15:16, 42.69it/s]




 22%|██▏       | 10868/50000 [02:57<14:38, 44.56it/s]




 22%|██▏       | 10875/50000 [02:57<13:04, 49.88it/s]




 22%|██▏       | 10881/50000 [02:57<12:57, 50.33it/s]




 22%|██▏       | 10887/50000 [02:57<12:46, 51.04it/s]




 22%|██▏       | 10894/50000 [02:57<11:55, 54.68it/s]




 22%|██▏       | 10901/50000 [02:57<11:25, 57.01it/s]




 22%|██▏       | 10907/50000 [02:57<11:28, 56.80it/s]




 22%|██▏       | 10913/50000 [02:57<11:44, 55.45it/s]




 22%|██▏       | 10919/50000 [02:58<11:39, 55.83it/s]




 22%|██▏       | 10927/50000 [02:58<10:54, 59.65it/s]




 22%|██▏       | 10934/50000 [02:58<10:38, 61.16it/s]




 22%|██▏       | 10942/50000 [02:58<09:58, 65.21it/s]




 22%|██▏       | 10949/50000 [02:58<09:52, 65.88it/s]




 22%|██▏       | 10956/50000 [02:58<10:16, 63.30it/s]




 22%|██▏       | 10963/50000 [02:58<10:18, 63.11it/s]




 22%|██▏       | 10970/50000 [02:58<10:33, 61.65it/s]




 22%|██▏       | 10977/50000 [02:58<10:19, 62.95it/s]




 22%|██▏       | 10984/50000 [02:59<11:13, 57.95it/s]




 22%|██▏       | 10990/50000 [02:59<11:48, 55.05it/s]




 22%|██▏       | 10996/50000 [02:59<11:32, 56.33it/s]




 22%|██▏       | 11003/50000 [02:59<11:03, 58.75it/s]




 22%|██▏       | 11010/50000 [02:59<10:40, 60.88it/s]




 22%|██▏       | 11018/50000 [02:59<10:12, 63.67it/s]




 22%|██▏       | 11025/50000 [02:59<10:23, 62.52it/s]




 22%|██▏       | 11032/50000 [02:59<10:07, 64.11it/s]




 22%|██▏       | 11039/50000 [02:59<09:53, 65.63it/s]




 22%|██▏       | 11046/50000 [03:00<09:51, 65.81it/s]




 22%|██▏       | 11053/50000 [03:00<10:12, 63.60it/s]




 22%|██▏       | 11061/50000 [03:00<09:42, 66.80it/s]




 22%|██▏       | 11068/50000 [03:00<10:16, 63.20it/s]




 22%|██▏       | 11076/50000 [03:00<09:51, 65.83it/s]




 22%|██▏       | 11083/50000 [03:00<10:04, 64.38it/s]




 22%|██▏       | 11091/50000 [03:00<09:41, 66.89it/s]




 22%|██▏       | 11099/50000 [03:00<09:22, 69.12it/s]




 22%|██▏       | 11106/50000 [03:00<10:04, 64.29it/s]




 22%|██▏       | 11113/50000 [03:01<10:14, 63.28it/s]




 22%|██▏       | 11121/50000 [03:01<09:54, 65.41it/s]




 22%|██▏       | 11128/50000 [03:01<09:57, 65.10it/s]




 22%|██▏       | 11136/50000 [03:01<09:34, 67.60it/s]




 22%|██▏       | 11143/50000 [03:01<10:16, 63.02it/s]




 22%|██▏       | 11151/50000 [03:01<09:45, 66.35it/s]




 22%|██▏       | 11159/50000 [03:01<09:32, 67.85it/s]




 22%|██▏       | 11166/50000 [03:01<09:38, 67.15it/s]




 22%|██▏       | 11174/50000 [03:01<09:33, 67.74it/s]




 22%|██▏       | 11181/50000 [03:02<09:55, 65.21it/s]




 22%|██▏       | 11189/50000 [03:02<09:39, 67.01it/s]




 22%|██▏       | 11196/50000 [03:02<09:36, 67.35it/s]




 22%|██▏       | 11203/50000 [03:02<09:56, 65.04it/s]




 22%|██▏       | 11210/50000 [03:02<10:05, 64.05it/s]




 22%|██▏       | 11218/50000 [03:02<09:34, 67.48it/s]




 22%|██▏       | 11225/50000 [03:02<09:57, 64.86it/s]




 22%|██▏       | 11233/50000 [03:02<09:49, 65.77it/s]




 22%|██▏       | 11240/50000 [03:02<09:46, 66.09it/s]




 22%|██▏       | 11247/50000 [03:03<10:12, 63.26it/s]




 23%|██▎       | 11254/50000 [03:03<10:25, 61.91it/s]




 23%|██▎       | 11261/50000 [03:03<10:11, 63.32it/s]




 23%|██▎       | 11269/50000 [03:03<09:43, 66.33it/s]




 23%|██▎       | 11277/50000 [03:03<09:19, 69.24it/s]




 23%|██▎       | 11285/50000 [03:03<09:24, 68.57it/s]




 23%|██▎       | 11292/50000 [03:03<09:53, 65.20it/s]




 23%|██▎       | 11300/50000 [03:03<09:36, 67.08it/s]




 23%|██▎       | 11307/50000 [03:03<09:38, 66.88it/s]




 23%|██▎       | 11314/50000 [03:04<10:01, 64.29it/s]




 23%|██▎       | 11322/50000 [03:04<09:27, 68.19it/s]




 23%|██▎       | 11329/50000 [03:04<09:37, 67.00it/s]




 23%|██▎       | 11338/50000 [03:04<09:02, 71.24it/s]




 23%|██▎       | 11346/50000 [03:04<09:13, 69.86it/s]




 23%|██▎       | 11354/50000 [03:04<09:40, 66.53it/s]




 23%|██▎       | 11361/50000 [03:04<10:24, 61.87it/s]




 23%|██▎       | 11368/50000 [03:04<10:30, 61.29it/s]




 23%|██▎       | 11376/50000 [03:05<09:58, 64.53it/s]




 23%|██▎       | 11383/50000 [03:05<09:48, 65.57it/s]




 23%|██▎       | 11391/50000 [03:05<09:32, 67.44it/s]




 23%|██▎       | 11398/50000 [03:05<09:58, 64.48it/s]




 23%|██▎       | 11406/50000 [03:05<09:39, 66.63it/s]




 23%|██▎       | 11413/50000 [03:05<09:41, 66.41it/s]




 23%|██▎       | 11421/50000 [03:05<09:19, 68.94it/s]




 23%|██▎       | 11428/50000 [03:05<09:34, 67.12it/s]




 23%|██▎       | 11435/50000 [03:05<09:35, 67.04it/s]




 23%|██▎       | 11442/50000 [03:06<10:38, 60.41it/s]




 23%|██▎       | 11449/50000 [03:06<11:14, 57.19it/s]




 23%|██▎       | 11455/50000 [03:06<11:13, 57.22it/s]




 23%|██▎       | 11463/50000 [03:06<11:45, 54.61it/s]




 23%|██▎       | 11470/50000 [03:06<11:15, 57.08it/s]




 23%|██▎       | 11476/50000 [03:06<11:44, 54.71it/s]




 23%|██▎       | 11484/50000 [03:06<10:46, 59.54it/s]




 23%|██▎       | 11492/50000 [03:06<09:58, 64.37it/s]




 23%|██▎       | 11500/50000 [03:06<09:26, 67.96it/s]




 23%|██▎       | 11508/50000 [03:07<09:04, 70.66it/s]




 23%|██▎       | 11516/50000 [03:07<09:29, 67.62it/s]




 23%|██▎       | 11523/50000 [03:07<10:08, 63.20it/s]




 23%|██▎       | 11530/50000 [03:07<10:00, 64.08it/s]




 23%|██▎       | 11537/50000 [03:07<10:55, 58.70it/s]




 23%|██▎       | 11544/50000 [03:07<11:18, 56.65it/s]




 23%|██▎       | 11553/50000 [03:07<10:06, 63.37it/s]




 23%|██▎       | 11560/50000 [03:07<10:38, 60.23it/s]




 23%|██▎       | 11567/50000 [03:08<10:52, 58.87it/s]




 23%|██▎       | 11574/50000 [03:08<11:13, 57.03it/s]




 23%|██▎       | 11582/50000 [03:08<10:16, 62.30it/s]




 23%|██▎       | 11590/50000 [03:08<09:48, 65.27it/s]




 23%|██▎       | 11597/50000 [03:08<10:26, 61.34it/s]




 23%|██▎       | 11604/50000 [03:08<11:06, 57.58it/s]




 23%|██▎       | 11611/50000 [03:08<10:38, 60.15it/s]




 23%|██▎       | 11620/50000 [03:08<09:44, 65.67it/s]




 23%|██▎       | 11628/50000 [03:09<09:18, 68.73it/s]




 23%|██▎       | 11636/50000 [03:09<09:01, 70.87it/s]




 23%|██▎       | 11644/50000 [03:09<09:55, 64.41it/s]




 23%|██▎       | 11651/50000 [03:09<10:06, 63.19it/s]




 23%|██▎       | 11658/50000 [03:09<10:42, 59.66it/s]




 23%|██▎       | 11665/50000 [03:09<11:43, 54.53it/s]




 23%|██▎       | 11671/50000 [03:09<13:47, 46.32it/s]




 23%|██▎       | 11676/50000 [03:09<15:00, 42.57it/s]




 23%|██▎       | 11681/50000 [03:10<14:56, 42.76it/s]




 23%|██▎       | 11686/50000 [03:10<15:12, 41.98it/s]




 23%|██▎       | 11691/50000 [03:10<14:45, 43.28it/s]




 23%|██▎       | 11696/50000 [03:10<14:21, 44.47it/s]




 23%|██▎       | 11702/50000 [03:10<13:36, 46.89it/s]




 23%|██▎       | 11707/50000 [03:10<14:45, 43.23it/s]




 23%|██▎       | 11713/50000 [03:10<13:51, 46.03it/s]




 23%|██▎       | 11718/50000 [03:10<14:49, 43.06it/s]




 23%|██▎       | 11723/50000 [03:11<15:15, 41.81it/s]




 23%|██▎       | 11729/50000 [03:11<14:31, 43.91it/s]




 23%|██▎       | 11734/50000 [03:11<15:21, 41.53it/s]




 23%|██▎       | 11741/50000 [03:11<13:37, 46.80it/s]




 23%|██▎       | 11746/50000 [03:11<13:46, 46.30it/s]




 24%|██▎       | 11752/50000 [03:11<13:08, 48.53it/s]




 24%|██▎       | 11758/50000 [03:11<12:51, 49.54it/s]




 24%|██▎       | 11764/50000 [03:11<12:41, 50.18it/s]




 24%|██▎       | 11770/50000 [03:11<12:04, 52.77it/s]




 24%|██▎       | 11776/50000 [03:12<12:41, 50.22it/s]




 24%|██▎       | 11783/50000 [03:12<11:49, 53.83it/s]




 24%|██▎       | 11789/50000 [03:12<13:08, 48.47it/s]




 24%|██▎       | 11795/50000 [03:12<12:35, 50.56it/s]




 24%|██▎       | 11801/50000 [03:12<13:17, 47.89it/s]




 24%|██▎       | 11806/50000 [03:12<14:22, 44.29it/s]




 24%|██▎       | 11813/50000 [03:12<13:04, 48.67it/s]




 24%|██▎       | 11819/50000 [03:12<13:22, 47.58it/s]




 24%|██▎       | 11824/50000 [03:13<13:48, 46.06it/s]




 24%|██▎       | 11829/50000 [03:13<14:18, 44.45it/s]




 24%|██▎       | 11834/50000 [03:13<14:48, 42.97it/s]




 24%|██▎       | 11840/50000 [03:13<13:50, 45.92it/s]




 24%|██▎       | 11848/50000 [03:13<12:08, 52.37it/s]




 24%|██▎       | 11855/50000 [03:13<11:14, 56.59it/s]




 24%|██▎       | 11862/50000 [03:13<10:56, 58.07it/s]




 24%|██▎       | 11869/50000 [03:13<11:02, 57.55it/s]




 24%|██▍       | 11878/50000 [03:14<10:06, 62.88it/s]




 24%|██▍       | 11885/50000 [03:14<09:58, 63.67it/s]




 24%|██▍       | 11893/50000 [03:14<09:48, 64.75it/s]




 24%|██▍       | 11900/50000 [03:14<10:36, 59.84it/s]




 24%|██▍       | 11908/50000 [03:14<09:59, 63.55it/s]




 24%|██▍       | 11917/50000 [03:14<09:20, 68.00it/s]




 24%|██▍       | 11925/50000 [03:14<09:21, 67.82it/s]




 24%|██▍       | 11933/50000 [03:14<09:18, 68.22it/s]




 24%|██▍       | 11940/50000 [03:14<10:13, 62.04it/s]




 24%|██▍       | 11947/50000 [03:15<11:00, 57.63it/s]




 24%|██▍       | 11953/50000 [03:15<11:51, 53.48it/s]




 24%|██▍       | 11959/50000 [03:15<12:15, 51.75it/s]




 24%|██▍       | 11965/50000 [03:15<13:51, 45.76it/s]




 24%|██▍       | 11971/50000 [03:15<13:43, 46.18it/s]




 24%|██▍       | 11976/50000 [03:15<13:37, 46.51it/s]




 24%|██▍       | 11985/50000 [03:15<11:44, 53.93it/s]




 24%|██▍       | 11991/50000 [03:15<11:36, 54.60it/s]




 24%|██▍       | 11997/50000 [03:16<12:00, 52.76it/s]




 24%|██▍       | 12003/50000 [03:16<11:41, 54.19it/s]




 24%|██▍       | 12012/50000 [03:16<10:34, 59.85it/s]




 24%|██▍       | 12019/50000 [03:16<10:35, 59.74it/s]




 24%|██▍       | 12026/50000 [03:16<11:11, 56.56it/s]




 24%|██▍       | 12034/50000 [03:16<10:16, 61.62it/s]




 24%|██▍       | 12043/50000 [03:16<09:35, 66.01it/s]




 24%|██▍       | 12051/50000 [03:16<09:08, 69.17it/s]




 24%|██▍       | 12060/50000 [03:16<08:42, 72.58it/s]




 24%|██▍       | 12068/50000 [03:17<09:21, 67.54it/s]




 24%|██▍       | 12076/50000 [03:17<09:09, 69.01it/s]




 24%|██▍       | 12084/50000 [03:17<09:10, 68.88it/s]




 24%|██▍       | 12091/50000 [03:17<09:46, 64.67it/s]




 24%|██▍       | 12098/50000 [03:17<10:14, 61.69it/s]




 24%|██▍       | 12105/50000 [03:17<09:56, 63.49it/s]




 24%|██▍       | 12113/50000 [03:17<09:30, 66.38it/s]




 24%|██▍       | 12120/50000 [03:17<09:37, 65.59it/s]




 24%|██▍       | 12128/50000 [03:18<09:25, 66.95it/s]




 24%|██▍       | 12135/50000 [03:18<09:57, 63.38it/s]




 24%|██▍       | 12142/50000 [03:18<10:20, 61.02it/s]




 24%|██▍       | 12149/50000 [03:18<10:00, 63.00it/s]




 24%|██▍       | 12157/50000 [03:18<09:31, 66.17it/s]




 24%|██▍       | 12166/50000 [03:18<08:58, 70.32it/s]




 24%|██▍       | 12174/50000 [03:18<09:04, 69.51it/s]




 24%|██▍       | 12183/50000 [03:18<08:38, 72.93it/s]




 24%|██▍       | 12193/50000 [03:18<08:03, 78.13it/s]




 24%|██▍       | 12201/50000 [03:19<08:16, 76.05it/s]




 24%|██▍       | 12209/50000 [03:19<08:17, 75.94it/s]




 24%|██▍       | 12218/50000 [03:19<08:00, 78.69it/s]




 24%|██▍       | 12226/50000 [03:19<08:11, 76.86it/s]




 24%|██▍       | 12234/50000 [03:19<08:37, 72.96it/s]




 24%|██▍       | 12242/50000 [03:19<08:29, 74.16it/s]




 25%|██▍       | 12251/50000 [03:19<08:13, 76.55it/s]




 25%|██▍       | 12259/50000 [03:19<08:16, 76.07it/s]




 25%|██▍       | 12267/50000 [03:19<08:13, 76.38it/s]




 25%|██▍       | 12275/50000 [03:20<08:12, 76.61it/s]




 25%|██▍       | 12283/50000 [03:20<08:07, 77.43it/s]




 25%|██▍       | 12292/50000 [03:20<07:50, 80.12it/s]




 25%|██▍       | 12301/50000 [03:20<08:12, 76.48it/s]




 25%|██▍       | 12309/50000 [03:20<08:36, 72.91it/s]




 25%|██▍       | 12317/50000 [03:20<08:24, 74.75it/s]




 25%|██▍       | 12326/50000 [03:20<08:11, 76.60it/s]




 25%|██▍       | 12337/50000 [03:20<07:39, 81.89it/s]




 25%|██▍       | 12346/50000 [03:20<07:57, 78.92it/s]




 25%|██▍       | 12356/50000 [03:21<07:43, 81.23it/s]




 25%|██▍       | 12365/50000 [03:21<07:54, 79.32it/s]




 25%|██▍       | 12374/50000 [03:21<07:44, 81.09it/s]




 25%|██▍       | 12383/50000 [03:21<07:30, 83.51it/s]




 25%|██▍       | 12392/50000 [03:21<07:31, 83.29it/s]




 25%|██▍       | 12401/50000 [03:21<07:28, 83.84it/s]




 25%|██▍       | 12411/50000 [03:21<07:24, 84.58it/s]




 25%|██▍       | 12420/50000 [03:21<07:55, 79.06it/s]




 25%|██▍       | 12429/50000 [03:21<07:50, 79.82it/s]




 25%|██▍       | 12438/50000 [03:22<07:47, 80.41it/s]




 25%|██▍       | 12447/50000 [03:22<07:56, 78.88it/s]




 25%|██▍       | 12456/50000 [03:22<07:44, 80.76it/s]




 25%|██▍       | 12465/50000 [03:22<07:40, 81.58it/s]




 25%|██▍       | 12474/50000 [03:22<08:07, 76.91it/s]




 25%|██▍       | 12483/50000 [03:22<07:52, 79.41it/s]




 25%|██▍       | 12492/50000 [03:22<08:05, 77.30it/s]




 25%|██▌       | 12500/50000 [03:22<09:31, 65.67it/s]




 25%|██▌       | 12507/50000 [03:23<09:34, 65.28it/s]




 25%|██▌       | 12516/50000 [03:23<09:06, 68.59it/s]




 25%|██▌       | 12524/50000 [03:23<09:30, 65.72it/s]




 25%|██▌       | 12531/50000 [03:23<09:48, 63.71it/s]




 25%|██▌       | 12540/50000 [03:23<08:57, 69.69it/s]




 25%|██▌       | 12548/50000 [03:23<09:01, 69.12it/s]




 25%|██▌       | 12556/50000 [03:23<09:12, 67.73it/s]




 25%|██▌       | 12563/50000 [03:23<09:22, 66.50it/s]




 25%|██▌       | 12570/50000 [03:23<09:45, 63.88it/s]




 25%|██▌       | 12577/50000 [03:24<09:58, 62.49it/s]




 25%|██▌       | 12584/50000 [03:24<09:56, 62.71it/s]




 25%|██▌       | 12593/50000 [03:24<09:20, 66.73it/s]




 25%|██▌       | 12600/50000 [03:24<09:16, 67.15it/s]




 25%|██▌       | 12607/50000 [03:24<10:00, 62.24it/s]




 25%|██▌       | 12614/50000 [03:24<09:53, 63.04it/s]




 25%|██▌       | 12622/50000 [03:24<09:24, 66.20it/s]




 25%|██▌       | 12629/50000 [03:24<09:45, 63.85it/s]




 25%|██▌       | 12636/50000 [03:24<09:56, 62.66it/s]




 25%|██▌       | 12643/50000 [03:25<10:04, 61.85it/s]




 25%|██▌       | 12650/50000 [03:25<11:01, 56.47it/s]




 25%|██▌       | 12657/50000 [03:25<10:28, 59.38it/s]




 25%|██▌       | 12665/50000 [03:25<10:18, 60.33it/s]




 25%|██▌       | 12672/50000 [03:25<11:44, 52.99it/s]




 25%|██▌       | 12679/50000 [03:25<11:06, 55.97it/s]




 25%|██▌       | 12686/50000 [03:25<10:42, 58.11it/s]




 25%|██▌       | 12692/50000 [03:25<10:55, 56.88it/s]




 25%|██▌       | 12698/50000 [03:26<11:18, 54.97it/s]




 25%|██▌       | 12704/50000 [03:26<11:34, 53.71it/s]




 25%|██▌       | 12710/50000 [03:26<12:48, 48.51it/s]




 25%|██▌       | 12715/50000 [03:26<13:10, 47.19it/s]




 25%|██▌       | 12722/50000 [03:26<12:22, 50.24it/s]




 25%|██▌       | 12728/50000 [03:26<11:47, 52.71it/s]




 25%|██▌       | 12735/50000 [03:26<11:04, 56.08it/s]




 25%|██▌       | 12743/50000 [03:26<10:32, 58.86it/s]




 26%|██▌       | 12752/50000 [03:27<09:36, 64.59it/s]




 26%|██▌       | 12761/50000 [03:27<08:49, 70.28it/s]




 26%|██▌       | 12769/50000 [03:27<08:34, 72.40it/s]




 26%|██▌       | 12777/50000 [03:27<08:39, 71.58it/s]




 26%|██▌       | 12785/50000 [03:27<11:03, 56.12it/s]




 26%|██▌       | 12792/50000 [03:27<10:47, 57.51it/s]




 26%|██▌       | 12801/50000 [03:27<09:53, 62.65it/s]




 26%|██▌       | 12809/50000 [03:27<09:20, 66.39it/s]




 26%|██▌       | 12817/50000 [03:28<09:17, 66.69it/s]




 26%|██▌       | 12824/50000 [03:28<09:17, 66.67it/s]




 26%|██▌       | 12831/50000 [03:28<09:12, 67.29it/s]




 26%|██▌       | 12838/50000 [03:28<09:35, 64.56it/s]




 26%|██▌       | 12845/50000 [03:28<09:50, 62.95it/s]




 26%|██▌       | 12852/50000 [03:28<09:44, 63.55it/s]




 26%|██▌       | 12859/50000 [03:28<09:46, 63.28it/s]




 26%|██▌       | 12866/50000 [03:28<09:35, 64.49it/s]




 26%|██▌       | 12874/50000 [03:28<09:08, 67.65it/s]




 26%|██▌       | 12883/50000 [03:28<08:33, 72.26it/s]




 26%|██▌       | 12891/50000 [03:29<08:30, 72.66it/s]




 26%|██▌       | 12899/50000 [03:29<08:46, 70.44it/s]




 26%|██▌       | 12908/50000 [03:29<08:25, 73.38it/s]




 26%|██▌       | 12917/50000 [03:29<08:00, 77.15it/s]




 26%|██▌       | 12925/50000 [03:29<08:20, 74.05it/s]




 26%|██▌       | 12933/50000 [03:29<08:30, 72.60it/s]




 26%|██▌       | 12941/50000 [03:29<08:42, 70.96it/s]




 26%|██▌       | 12949/50000 [03:29<08:33, 72.11it/s]




 26%|██▌       | 12957/50000 [03:30<09:32, 64.66it/s]




 26%|██▌       | 12964/50000 [03:30<10:35, 58.30it/s]




 26%|██▌       | 12971/50000 [03:30<11:06, 55.58it/s]




 26%|██▌       | 12979/50000 [03:30<10:22, 59.45it/s]




 26%|██▌       | 12987/50000 [03:30<09:37, 64.07it/s]




 26%|██▌       | 12995/50000 [03:30<09:10, 67.16it/s]




 26%|██▌       | 13004/50000 [03:30<08:40, 71.02it/s]




 26%|██▌       | 13012/50000 [03:30<08:50, 69.72it/s]




 26%|██▌       | 13021/50000 [03:30<08:19, 74.08it/s]




 26%|██▌       | 13029/50000 [03:31<08:31, 72.33it/s]




 26%|██▌       | 13037/50000 [03:31<09:20, 65.90it/s]




 26%|██▌       | 13044/50000 [03:31<09:59, 61.64it/s]




 26%|██▌       | 13051/50000 [03:31<10:13, 60.20it/s]




 26%|██▌       | 13058/50000 [03:31<10:01, 61.40it/s]




 26%|██▌       | 13066/50000 [03:31<09:27, 65.08it/s]




 26%|██▌       | 13074/50000 [03:31<09:06, 67.59it/s]




 26%|██▌       | 13081/50000 [03:31<09:19, 66.03it/s]




 26%|██▌       | 13088/50000 [03:32<09:48, 62.72it/s]




 26%|██▌       | 13095/50000 [03:32<10:26, 58.92it/s]




 26%|██▌       | 13102/50000 [03:32<09:57, 61.74it/s]




 26%|██▌       | 13109/50000 [03:32<09:43, 63.19it/s]




 26%|██▌       | 13117/50000 [03:32<09:35, 64.09it/s]




 26%|██▋       | 13126/50000 [03:32<08:47, 69.86it/s]




 26%|██▋       | 13134/50000 [03:32<08:43, 70.49it/s]




 26%|██▋       | 13142/50000 [03:32<09:10, 67.00it/s]




 26%|██▋       | 13149/50000 [03:33<10:32, 58.28it/s]




 26%|██▋       | 13156/50000 [03:33<10:40, 57.48it/s]




 26%|██▋       | 13162/50000 [03:33<11:25, 53.73it/s]




 26%|██▋       | 13169/50000 [03:33<10:51, 56.54it/s]




 26%|██▋       | 13176/50000 [03:33<10:32, 58.25it/s]




 26%|██▋       | 13185/50000 [03:33<09:39, 63.53it/s]




 26%|██▋       | 13192/50000 [03:33<09:26, 64.94it/s]




 26%|██▋       | 13200/50000 [03:33<09:16, 66.14it/s]




 26%|██▋       | 13208/50000 [03:33<08:51, 69.27it/s]




 26%|██▋       | 13216/50000 [03:34<09:08, 67.07it/s]




 26%|██▋       | 13223/50000 [03:34<09:42, 63.11it/s]




 26%|██▋       | 13230/50000 [03:34<10:29, 58.42it/s]




 26%|██▋       | 13237/50000 [03:34<10:25, 58.73it/s]




 26%|██▋       | 13243/50000 [03:34<10:28, 58.46it/s]




 26%|██▋       | 13250/50000 [03:34<09:58, 61.39it/s]




 27%|██▋       | 13258/50000 [03:34<09:21, 65.39it/s]




 27%|██▋       | 13267/50000 [03:34<08:51, 69.15it/s]




 27%|██▋       | 13276/50000 [03:34<08:19, 73.46it/s]




 27%|██▋       | 13284/50000 [03:35<08:40, 70.52it/s]




 27%|██▋       | 13292/50000 [03:35<09:41, 63.15it/s]




 27%|██▋       | 13299/50000 [03:35<09:56, 61.51it/s]




 27%|██▋       | 13306/50000 [03:35<09:57, 61.37it/s]




 27%|██▋       | 13315/50000 [03:35<09:00, 67.83it/s]




 27%|██▋       | 13324/50000 [03:35<08:24, 72.69it/s]




 27%|██▋       | 13332/50000 [03:35<08:47, 69.55it/s]




 27%|██▋       | 13340/50000 [03:35<08:34, 71.28it/s]




 27%|██▋       | 13348/50000 [03:36<08:54, 68.63it/s]




 27%|██▋       | 13356/50000 [03:36<09:11, 66.39it/s]




 27%|██▋       | 13363/50000 [03:36<09:38, 63.29it/s]




 27%|██▋       | 13370/50000 [03:36<09:27, 64.49it/s]




 27%|██▋       | 13377/50000 [03:36<09:32, 63.93it/s]




 27%|██▋       | 13384/50000 [03:36<09:18, 65.51it/s]




 27%|██▋       | 13393/50000 [03:36<08:48, 69.29it/s]




 27%|██▋       | 13401/50000 [03:36<10:29, 58.11it/s]




 27%|██▋       | 13408/50000 [03:37<11:12, 54.43it/s]




 27%|██▋       | 13414/50000 [03:37<11:08, 54.71it/s]




 27%|██▋       | 13420/50000 [03:37<11:01, 55.31it/s]




 27%|██▋       | 13426/50000 [03:37<11:10, 54.52it/s]




 27%|██▋       | 13432/50000 [03:37<10:57, 55.63it/s]




 27%|██▋       | 13438/50000 [03:37<11:18, 53.86it/s]




 27%|██▋       | 13445/50000 [03:37<10:36, 57.47it/s]




 27%|██▋       | 13452/50000 [03:37<10:18, 59.09it/s]




 27%|██▋       | 13459/50000 [03:37<10:03, 60.59it/s]




 27%|██▋       | 13466/50000 [03:38<10:40, 57.02it/s]




 27%|██▋       | 13473/50000 [03:38<10:12, 59.65it/s]




 27%|██▋       | 13480/50000 [03:38<10:12, 59.60it/s]




 27%|██▋       | 13487/50000 [03:38<10:05, 60.33it/s]




 27%|██▋       | 13494/50000 [03:38<10:35, 57.42it/s]




 27%|██▋       | 13501/50000 [03:38<10:12, 59.56it/s]




 27%|██▋       | 13508/50000 [03:38<10:31, 57.77it/s]




 27%|██▋       | 13514/50000 [03:38<10:43, 56.74it/s]




 27%|██▋       | 13520/50000 [03:39<11:24, 53.28it/s]




 27%|██▋       | 13526/50000 [03:39<12:08, 50.07it/s]




 27%|██▋       | 13532/50000 [03:39<12:35, 48.28it/s]




 27%|██▋       | 13537/50000 [03:39<12:37, 48.12it/s]




 27%|██▋       | 13542/50000 [03:39<12:48, 47.46it/s]




 27%|██▋       | 13547/50000 [03:39<12:48, 47.41it/s]




 27%|██▋       | 13554/50000 [03:39<12:08, 50.05it/s]




 27%|██▋       | 13560/50000 [03:39<11:59, 50.65it/s]




 27%|██▋       | 13566/50000 [03:39<12:07, 50.05it/s]




 27%|██▋       | 13572/50000 [03:40<11:47, 51.49it/s]




 27%|██▋       | 13578/50000 [03:40<12:11, 49.81it/s]




 27%|██▋       | 13584/50000 [03:40<11:45, 51.62it/s]




 27%|██▋       | 13591/50000 [03:40<11:04, 54.81it/s]




 27%|██▋       | 13599/50000 [03:40<10:09, 59.68it/s]




 27%|██▋       | 13606/50000 [03:40<10:08, 59.77it/s]




 27%|██▋       | 13613/50000 [03:40<10:57, 55.31it/s]




 27%|██▋       | 13620/50000 [03:40<10:42, 56.64it/s]




 27%|██▋       | 13627/50000 [03:41<10:15, 59.06it/s]




 27%|██▋       | 13635/50000 [03:41<09:42, 62.48it/s]




 27%|██▋       | 13642/50000 [03:41<09:42, 62.37it/s]




 27%|██▋       | 13649/50000 [03:41<09:57, 60.84it/s]




 27%|██▋       | 13657/50000 [03:41<09:38, 62.82it/s]




 27%|██▋       | 13664/50000 [03:41<11:03, 54.76it/s]




 27%|██▋       | 13670/50000 [03:41<11:23, 53.14it/s]




 27%|██▋       | 13676/50000 [03:41<11:44, 51.53it/s]




 27%|██▋       | 13682/50000 [03:42<11:35, 52.21it/s]




 27%|██▋       | 13688/50000 [03:42<12:05, 50.08it/s]




 27%|██▋       | 13694/50000 [03:42<12:45, 47.43it/s]




 27%|██▋       | 13699/50000 [03:42<13:08, 46.02it/s]




 27%|██▋       | 13704/50000 [03:42<14:33, 41.55it/s]




 27%|██▋       | 13709/50000 [03:42<15:26, 39.18it/s]




 27%|██▋       | 13714/50000 [03:42<14:47, 40.90it/s]




 27%|██▋       | 13719/50000 [03:42<14:13, 42.52it/s]




 27%|██▋       | 13724/50000 [03:43<14:04, 42.95it/s]




 27%|██▋       | 13730/50000 [03:43<13:26, 44.98it/s]




 27%|██▋       | 13738/50000 [03:43<11:53, 50.82it/s]




 27%|██▋       | 13744/50000 [03:43<11:25, 52.89it/s]




 28%|██▊       | 13750/50000 [03:43<11:51, 50.97it/s]




 28%|██▊       | 13756/50000 [03:43<11:27, 52.72it/s]




 28%|██▊       | 13764/50000 [03:43<10:31, 57.38it/s]




 28%|██▊       | 13771/50000 [03:43<10:01, 60.21it/s]




 28%|██▊       | 13778/50000 [03:43<10:20, 58.35it/s]




 28%|██▊       | 13784/50000 [03:44<10:24, 58.03it/s]




 28%|██▊       | 13790/50000 [03:44<10:49, 55.72it/s]




 28%|██▊       | 13796/50000 [03:44<10:43, 56.28it/s]




 28%|██▊       | 13802/50000 [03:44<11:27, 52.64it/s]




 28%|██▊       | 13808/50000 [03:44<12:57, 46.57it/s]




 28%|██▊       | 13813/50000 [03:44<14:02, 42.94it/s]




 28%|██▊       | 13818/50000 [03:44<14:06, 42.74it/s]




 28%|██▊       | 13823/50000 [03:44<15:47, 38.18it/s]




 28%|██▊       | 13828/50000 [03:45<15:22, 39.19it/s]




 28%|██▊       | 13834/50000 [03:45<14:11, 42.47it/s]




 28%|██▊       | 13840/50000 [03:45<13:09, 45.78it/s]




 28%|██▊       | 13847/50000 [03:45<12:01, 50.08it/s]




 28%|██▊       | 13853/50000 [03:45<12:08, 49.59it/s]




 28%|██▊       | 13859/50000 [03:45<12:26, 48.43it/s]




 28%|██▊       | 13867/50000 [03:45<10:58, 54.85it/s]




 28%|██▊       | 13873/50000 [03:45<10:44, 56.03it/s]




 28%|██▊       | 13879/50000 [03:45<10:36, 56.72it/s]




 28%|██▊       | 13885/50000 [03:46<11:43, 51.36it/s]




 28%|██▊       | 13892/50000 [03:46<10:51, 55.45it/s]




 28%|██▊       | 13898/50000 [03:46<11:31, 52.20it/s]




 28%|██▊       | 13904/50000 [03:46<11:31, 52.24it/s]




 28%|██▊       | 13910/50000 [03:46<11:28, 52.40it/s]




 28%|██▊       | 13916/50000 [03:46<11:30, 52.24it/s]




 28%|██▊       | 13922/50000 [03:46<11:17, 53.23it/s]




 28%|██▊       | 13928/50000 [03:46<11:18, 53.14it/s]




 28%|██▊       | 13935/50000 [03:47<10:58, 54.76it/s]




 28%|██▊       | 13941/50000 [03:47<12:18, 48.81it/s]




 28%|██▊       | 13947/50000 [03:47<11:54, 50.44it/s]




 28%|██▊       | 13954/50000 [03:47<10:55, 54.96it/s]




 28%|██▊       | 13960/50000 [03:47<11:25, 52.58it/s]




 28%|██▊       | 13966/50000 [03:47<11:13, 53.48it/s]




 28%|██▊       | 13972/50000 [03:47<11:50, 50.70it/s]




 28%|██▊       | 13978/50000 [03:47<11:18, 53.07it/s]




 28%|██▊       | 13984/50000 [03:47<11:26, 52.43it/s]




 28%|██▊       | 13990/50000 [03:48<12:18, 48.76it/s]




 28%|██▊       | 13996/50000 [03:48<12:21, 48.53it/s]




 28%|██▊       | 14001/50000 [03:48<12:26, 48.21it/s]




 28%|██▊       | 14009/50000 [03:48<11:09, 53.77it/s]




 28%|██▊       | 14015/50000 [03:48<11:47, 50.89it/s]




 28%|██▊       | 14021/50000 [03:48<11:23, 52.66it/s]




 28%|██▊       | 14027/50000 [03:48<11:18, 53.04it/s]




 28%|██▊       | 14035/50000 [03:48<10:19, 58.08it/s]




 28%|██▊       | 14042/50000 [03:49<09:50, 60.93it/s]




 28%|██▊       | 14049/50000 [03:49<09:48, 61.12it/s]




 28%|██▊       | 14057/50000 [03:49<09:10, 65.34it/s]




 28%|██▊       | 14064/50000 [03:49<09:20, 64.16it/s]




 28%|██▊       | 14071/50000 [03:49<09:31, 62.85it/s]




 28%|██▊       | 14078/50000 [03:49<09:36, 62.29it/s]




 28%|██▊       | 14085/50000 [03:49<10:03, 59.55it/s]




 28%|██▊       | 14092/50000 [03:49<09:58, 59.98it/s]




 28%|██▊       | 14100/50000 [03:49<09:32, 62.75it/s]




 28%|██▊       | 14107/50000 [03:50<09:31, 62.79it/s]




 28%|██▊       | 14115/50000 [03:50<09:02, 66.17it/s]




 28%|██▊       | 14122/50000 [03:50<09:12, 64.90it/s]




 28%|██▊       | 14129/50000 [03:50<09:01, 66.20it/s]




 28%|██▊       | 14137/50000 [03:50<08:41, 68.78it/s]




 28%|██▊       | 14144/50000 [03:50<09:30, 62.88it/s]




 28%|██▊       | 14151/50000 [03:50<09:27, 63.15it/s]




 28%|██▊       | 14160/50000 [03:50<08:48, 67.84it/s]




 28%|██▊       | 14167/50000 [03:50<08:49, 67.73it/s]




 28%|██▊       | 14174/50000 [03:51<09:04, 65.76it/s]




 28%|██▊       | 14181/50000 [03:51<09:03, 65.89it/s]




 28%|██▊       | 14188/50000 [03:51<09:13, 64.71it/s]




 28%|██▊       | 14195/50000 [03:51<09:27, 63.05it/s]




 28%|██▊       | 14202/50000 [03:51<09:40, 61.64it/s]




 28%|██▊       | 14209/50000 [03:51<09:25, 63.29it/s]




 28%|██▊       | 14216/50000 [03:51<10:00, 59.56it/s]




 28%|██▊       | 14223/50000 [03:51<10:19, 57.77it/s]




 28%|██▊       | 14231/50000 [03:51<09:36, 62.04it/s]




 28%|██▊       | 14238/50000 [03:52<09:51, 60.46it/s]




 28%|██▊       | 14247/50000 [03:52<09:02, 65.92it/s]




 29%|██▊       | 14254/50000 [03:52<09:25, 63.16it/s]




 29%|██▊       | 14261/50000 [03:52<09:43, 61.20it/s]




 29%|██▊       | 14269/50000 [03:52<09:09, 65.08it/s]




 29%|██▊       | 14276/50000 [03:52<09:32, 62.45it/s]




 29%|██▊       | 14283/50000 [03:52<09:35, 62.02it/s]




 29%|██▊       | 14290/50000 [03:52<10:22, 57.40it/s]




 29%|██▊       | 14296/50000 [03:53<10:20, 57.53it/s]




 29%|██▊       | 14303/50000 [03:53<09:56, 59.89it/s]




 29%|██▊       | 14310/50000 [03:53<09:38, 61.66it/s]




 29%|██▊       | 14317/50000 [03:53<09:51, 60.28it/s]




 29%|██▊       | 14324/50000 [03:53<09:29, 62.70it/s]




 29%|██▊       | 14333/50000 [03:53<08:46, 67.77it/s]




 29%|██▊       | 14340/50000 [03:53<08:57, 66.34it/s]




 29%|██▊       | 14347/50000 [03:53<08:53, 66.87it/s]




 29%|██▊       | 14354/50000 [03:53<09:40, 61.42it/s]




 29%|██▊       | 14361/50000 [03:54<09:26, 62.95it/s]




 29%|██▊       | 14368/50000 [03:54<09:14, 64.25it/s]




 29%|██▉       | 14375/50000 [03:54<09:17, 63.94it/s]




 29%|██▉       | 14382/50000 [03:54<09:05, 65.33it/s]




 29%|██▉       | 14389/50000 [03:54<09:02, 65.59it/s]




 29%|██▉       | 14396/50000 [03:54<09:23, 63.23it/s]




 29%|██▉       | 14403/50000 [03:54<09:18, 63.74it/s]




 29%|██▉       | 14410/50000 [03:54<09:22, 63.25it/s]




 29%|██▉       | 14417/50000 [03:54<09:40, 61.35it/s]




 29%|██▉       | 14425/50000 [03:55<09:02, 65.60it/s]




 29%|██▉       | 14432/50000 [03:55<08:56, 66.24it/s]




 29%|██▉       | 14440/50000 [03:55<08:42, 68.12it/s]




 29%|██▉       | 14448/50000 [03:55<08:27, 70.04it/s]




 29%|██▉       | 14456/50000 [03:55<08:20, 71.07it/s]




 29%|██▉       | 14464/50000 [03:55<08:46, 67.46it/s]




 29%|██▉       | 14471/50000 [03:55<09:23, 63.10it/s]




 29%|██▉       | 14478/50000 [03:55<09:16, 63.83it/s]




 29%|██▉       | 14486/50000 [03:55<08:43, 67.82it/s]




 29%|██▉       | 14493/50000 [03:56<08:39, 68.41it/s]




 29%|██▉       | 14500/50000 [03:56<08:56, 66.20it/s]




 29%|██▉       | 14507/50000 [03:56<09:32, 62.00it/s]




 29%|██▉       | 14514/50000 [03:56<09:30, 62.20it/s]




 29%|██▉       | 14521/50000 [03:56<09:55, 59.56it/s]




 29%|██▉       | 14528/50000 [03:56<10:01, 59.01it/s]




 29%|██▉       | 14534/50000 [03:56<10:40, 55.33it/s]




 29%|██▉       | 14540/50000 [03:56<10:44, 54.99it/s]




 29%|██▉       | 14546/50000 [03:56<10:56, 54.01it/s]




 29%|██▉       | 14553/50000 [03:57<10:19, 57.17it/s]




 29%|██▉       | 14559/50000 [03:57<10:33, 55.93it/s]




 29%|██▉       | 14568/50000 [03:57<09:29, 62.22it/s]




 29%|██▉       | 14575/50000 [03:57<09:29, 62.19it/s]




 29%|██▉       | 14582/50000 [03:57<10:39, 55.39it/s]




 29%|██▉       | 14588/50000 [03:57<11:13, 52.61it/s]




 29%|██▉       | 14594/50000 [03:57<11:46, 50.09it/s]




 29%|██▉       | 14600/50000 [03:57<12:00, 49.12it/s]




 29%|██▉       | 14606/50000 [03:58<11:27, 51.45it/s]




 29%|██▉       | 14612/50000 [03:58<11:38, 50.66it/s]




 29%|██▉       | 14618/50000 [03:58<12:14, 48.20it/s]




 29%|██▉       | 14623/50000 [03:58<12:10, 48.41it/s]




 29%|██▉       | 14628/50000 [03:58<12:14, 48.13it/s]




 29%|██▉       | 14633/50000 [03:58<13:03, 45.16it/s]




 29%|██▉       | 14638/50000 [03:58<13:00, 45.29it/s]




 29%|██▉       | 14644/50000 [03:58<12:20, 47.75it/s]




 29%|██▉       | 14649/50000 [03:59<13:02, 45.17it/s]




 29%|██▉       | 14654/50000 [03:59<13:40, 43.07it/s]




 29%|██▉       | 14659/50000 [03:59<13:07, 44.85it/s]




 29%|██▉       | 14667/50000 [03:59<11:31, 51.09it/s]




 29%|██▉       | 14674/50000 [03:59<10:47, 54.53it/s]




 29%|██▉       | 14681/50000 [03:59<10:05, 58.30it/s]




 29%|██▉       | 14690/50000 [03:59<09:18, 63.23it/s]




 29%|██▉       | 14697/50000 [03:59<09:24, 62.55it/s]




 29%|██▉       | 14705/50000 [03:59<08:53, 66.15it/s]




 29%|██▉       | 14712/50000 [04:00<09:38, 60.99it/s]




 29%|██▉       | 14719/50000 [04:00<09:36, 61.16it/s]




 29%|██▉       | 14726/50000 [04:00<10:36, 55.40it/s]




 29%|██▉       | 14732/50000 [04:00<11:14, 52.30it/s]




 29%|██▉       | 14738/50000 [04:00<11:49, 49.71it/s]




 29%|██▉       | 14744/50000 [04:00<11:59, 48.99it/s]




 30%|██▉       | 14750/50000 [04:00<13:40, 42.98it/s]




 30%|██▉       | 14755/50000 [04:00<13:36, 43.16it/s]




 30%|██▉       | 14760/50000 [04:01<13:04, 44.92it/s]




 30%|██▉       | 14766/50000 [04:01<12:32, 46.80it/s]




 30%|██▉       | 14771/50000 [04:01<12:26, 47.21it/s]




 30%|██▉       | 14778/50000 [04:01<11:23, 51.51it/s]




 30%|██▉       | 14784/50000 [04:01<11:40, 50.26it/s]




 30%|██▉       | 14790/50000 [04:01<11:21, 51.64it/s]




 30%|██▉       | 14796/50000 [04:01<10:57, 53.50it/s]




 30%|██▉       | 14803/50000 [04:01<10:12, 57.47it/s]




 30%|██▉       | 14809/50000 [04:01<10:37, 55.20it/s]




 30%|██▉       | 14815/50000 [04:02<10:35, 55.35it/s]




 30%|██▉       | 14821/50000 [04:02<10:34, 55.46it/s]




 30%|██▉       | 14827/50000 [04:02<10:21, 56.63it/s]




 30%|██▉       | 14834/50000 [04:02<10:02, 58.39it/s]




 30%|██▉       | 14840/50000 [04:02<10:17, 56.98it/s]




 30%|██▉       | 14847/50000 [04:02<09:43, 60.23it/s]




 30%|██▉       | 14854/50000 [04:02<09:44, 60.16it/s]




 30%|██▉       | 14861/50000 [04:02<09:34, 61.21it/s]




 30%|██▉       | 14868/50000 [04:02<09:35, 61.00it/s]




 30%|██▉       | 14875/50000 [04:03<10:00, 58.49it/s]




 30%|██▉       | 14881/50000 [04:03<10:01, 58.43it/s]




 30%|██▉       | 14888/50000 [04:03<09:35, 61.04it/s]




 30%|██▉       | 14895/50000 [04:03<09:17, 63.01it/s]




 30%|██▉       | 14902/50000 [04:03<09:05, 64.29it/s]




 30%|██▉       | 14910/50000 [04:03<08:41, 67.33it/s]




 30%|██▉       | 14917/50000 [04:03<08:57, 65.31it/s]




 30%|██▉       | 14924/50000 [04:03<08:57, 65.21it/s]




 30%|██▉       | 14931/50000 [04:03<09:38, 60.59it/s]




 30%|██▉       | 14938/50000 [04:04<09:37, 60.72it/s]




 30%|██▉       | 14946/50000 [04:04<09:06, 64.16it/s]




 30%|██▉       | 14953/50000 [04:04<09:14, 63.25it/s]




 30%|██▉       | 14960/50000 [04:04<09:31, 61.26it/s]




 30%|██▉       | 14967/50000 [04:04<09:25, 62.00it/s]




 30%|██▉       | 14974/50000 [04:04<09:20, 62.53it/s]




 30%|██▉       | 14981/50000 [04:04<09:32, 61.18it/s]




 30%|██▉       | 14988/50000 [04:04<09:27, 61.70it/s]




 30%|██▉       | 14995/50000 [04:04<09:53, 58.94it/s]




 30%|███       | 15002/50000 [04:05<09:29, 61.43it/s]




 30%|███       | 15009/50000 [04:05<09:33, 60.99it/s]




 30%|███       | 15017/50000 [04:05<09:00, 64.76it/s]




 30%|███       | 15024/50000 [04:05<09:15, 62.91it/s]




 30%|███       | 15031/50000 [04:05<09:13, 63.18it/s]




 30%|███       | 15039/50000 [04:05<08:48, 66.14it/s]




 30%|███       | 15046/50000 [04:05<08:45, 66.54it/s]




 30%|███       | 15053/50000 [04:05<09:06, 63.99it/s]




 30%|███       | 15060/50000 [04:05<09:02, 64.46it/s]




 30%|███       | 15068/50000 [04:06<08:43, 66.67it/s]




 30%|███       | 15076/50000 [04:06<08:25, 69.03it/s]




 30%|███       | 15083/50000 [04:06<08:46, 66.26it/s]




 30%|███       | 15090/50000 [04:06<08:45, 66.43it/s]




 30%|███       | 15097/50000 [04:06<08:47, 66.18it/s]




 30%|███       | 15104/50000 [04:06<09:12, 63.15it/s]




 30%|███       | 15111/50000 [04:06<09:03, 64.22it/s]




 30%|███       | 15118/50000 [04:06<08:53, 65.35it/s]




 30%|███       | 15125/50000 [04:06<08:48, 65.97it/s]




 30%|███       | 15132/50000 [04:07<08:58, 64.77it/s]




 30%|███       | 15139/50000 [04:07<08:53, 65.38it/s]




 30%|███       | 15146/50000 [04:07<09:38, 60.30it/s]




 30%|███       | 15154/50000 [04:07<09:03, 64.06it/s]




 30%|███       | 15163/50000 [04:07<08:23, 69.20it/s]




 30%|███       | 15171/50000 [04:07<08:13, 70.64it/s]




 30%|███       | 15179/50000 [04:07<08:35, 67.53it/s]




 30%|███       | 15187/50000 [04:07<08:33, 67.84it/s]




 30%|███       | 15194/50000 [04:08<08:59, 64.55it/s]




 30%|███       | 15202/50000 [04:08<08:43, 66.43it/s]




 30%|███       | 15209/50000 [04:08<08:53, 65.26it/s]




 30%|███       | 15216/50000 [04:08<08:43, 66.47it/s]




 30%|███       | 15224/50000 [04:08<08:33, 67.77it/s]




 30%|███       | 15231/50000 [04:08<09:03, 63.98it/s]




 30%|███       | 15238/50000 [04:08<09:26, 61.41it/s]




 30%|███       | 15246/50000 [04:08<08:57, 64.62it/s]




 31%|███       | 15253/50000 [04:08<09:07, 63.50it/s]




 31%|███       | 15260/50000 [04:09<09:15, 62.57it/s]




 31%|███       | 15267/50000 [04:09<09:30, 60.88it/s]




 31%|███       | 15274/50000 [04:09<09:37, 60.16it/s]




 31%|███       | 15281/50000 [04:09<09:55, 58.31it/s]




 31%|███       | 15288/50000 [04:09<09:34, 60.39it/s]




 31%|███       | 15295/50000 [04:09<09:29, 60.95it/s]




 31%|███       | 15302/50000 [04:09<09:42, 59.57it/s]




 31%|███       | 15310/50000 [04:09<09:10, 63.04it/s]




 31%|███       | 15317/50000 [04:09<09:12, 62.76it/s]




 31%|███       | 15324/50000 [04:10<10:23, 55.58it/s]




 31%|███       | 15330/50000 [04:10<10:40, 54.11it/s]




 31%|███       | 15336/50000 [04:10<10:36, 54.43it/s]




 31%|███       | 15342/50000 [04:10<11:38, 49.65it/s]




 31%|███       | 15348/50000 [04:10<11:09, 51.79it/s]




 31%|███       | 15354/50000 [04:10<11:00, 52.49it/s]




 31%|███       | 15361/50000 [04:10<10:20, 55.83it/s]




 31%|███       | 15368/50000 [04:10<09:43, 59.33it/s]




 31%|███       | 15376/50000 [04:11<09:16, 62.26it/s]




 31%|███       | 15383/50000 [04:11<09:41, 59.52it/s]




 31%|███       | 15390/50000 [04:11<10:02, 57.47it/s]




 31%|███       | 15397/50000 [04:11<09:44, 59.23it/s]




 31%|███       | 15404/50000 [04:11<11:34, 49.83it/s]




 31%|███       | 15410/50000 [04:11<14:33, 39.58it/s]




 31%|███       | 15415/50000 [04:11<13:51, 41.61it/s]




 31%|███       | 15423/50000 [04:12<12:01, 47.94it/s]




 31%|███       | 15429/50000 [04:12<11:39, 49.41it/s]




 31%|███       | 15435/50000 [04:12<11:33, 49.87it/s]




 31%|███       | 15441/50000 [04:12<11:23, 50.58it/s]




 31%|███       | 15447/50000 [04:12<10:54, 52.77it/s]




 31%|███       | 15453/50000 [04:12<11:20, 50.77it/s]




 31%|███       | 15459/50000 [04:12<11:43, 49.09it/s]




 31%|███       | 15465/50000 [04:12<12:08, 47.41it/s]




 31%|███       | 15470/50000 [04:12<12:11, 47.18it/s]




 31%|███       | 15475/50000 [04:13<12:30, 45.98it/s]




 31%|███       | 15482/50000 [04:13<11:26, 50.29it/s]




 31%|███       | 15489/50000 [04:13<10:49, 53.09it/s]




 31%|███       | 15496/50000 [04:13<10:14, 56.18it/s]




 31%|███       | 15502/50000 [04:13<11:02, 52.10it/s]




 31%|███       | 15508/50000 [04:13<11:11, 51.37it/s]




 31%|███       | 15514/50000 [04:13<11:21, 50.60it/s]




 31%|███       | 15520/50000 [04:13<11:43, 48.98it/s]




 31%|███       | 15526/50000 [04:14<11:35, 49.57it/s]




 31%|███       | 15532/50000 [04:14<11:17, 50.87it/s]




 31%|███       | 15538/50000 [04:14<11:44, 48.92it/s]




 31%|███       | 15543/50000 [04:14<11:49, 48.56it/s]




 31%|███       | 15548/50000 [04:14<11:43, 48.95it/s]




 31%|███       | 15553/50000 [04:14<12:02, 47.68it/s]




 31%|███       | 15558/50000 [04:14<12:21, 46.44it/s]




 31%|███       | 15563/50000 [04:14<12:23, 46.30it/s]




 31%|███       | 15568/50000 [04:14<13:31, 42.45it/s]




 31%|███       | 15573/50000 [04:15<13:01, 44.03it/s]




 31%|███       | 15579/50000 [04:15<12:26, 46.12it/s]




 31%|███       | 15585/50000 [04:15<11:45, 48.75it/s]




 31%|███       | 15591/50000 [04:15<11:07, 51.56it/s]




 31%|███       | 15599/50000 [04:15<10:22, 55.24it/s]




 31%|███       | 15605/50000 [04:15<10:35, 54.11it/s]




 31%|███       | 15612/50000 [04:15<09:59, 57.32it/s]




 31%|███       | 15618/50000 [04:15<10:05, 56.82it/s]




 31%|███▏      | 15626/50000 [04:15<09:26, 60.71it/s]




 31%|███▏      | 15633/50000 [04:16<09:35, 59.72it/s]




 31%|███▏      | 15640/50000 [04:16<09:42, 58.97it/s]




 31%|███▏      | 15646/50000 [04:16<09:50, 58.15it/s]




 31%|███▏      | 15652/50000 [04:16<09:48, 58.39it/s]




 31%|███▏      | 15658/50000 [04:16<10:07, 56.50it/s]




 31%|███▏      | 15664/50000 [04:16<10:54, 52.48it/s]




 31%|███▏      | 15670/50000 [04:16<11:15, 50.83it/s]




 31%|███▏      | 15676/50000 [04:16<11:43, 48.78it/s]




 31%|███▏      | 15682/50000 [04:17<11:07, 51.45it/s]




 31%|███▏      | 15689/50000 [04:17<10:32, 54.24it/s]




 31%|███▏      | 15697/50000 [04:17<09:56, 57.49it/s]




 31%|███▏      | 15704/50000 [04:17<09:32, 59.86it/s]




 31%|███▏      | 15711/50000 [04:17<09:17, 61.47it/s]




 31%|███▏      | 15718/50000 [04:17<10:10, 56.18it/s]




 31%|███▏      | 15725/50000 [04:17<09:37, 59.31it/s]




 31%|███▏      | 15732/50000 [04:17<09:16, 61.54it/s]




 31%|███▏      | 15739/50000 [04:17<09:03, 63.04it/s]




 31%|███▏      | 15747/50000 [04:18<08:38, 66.04it/s]




 32%|███▏      | 15754/50000 [04:18<09:04, 62.89it/s]




 32%|███▏      | 15761/50000 [04:18<09:37, 59.32it/s]




 32%|███▏      | 15768/50000 [04:18<09:19, 61.23it/s]




 32%|███▏      | 15776/50000 [04:18<08:49, 64.64it/s]




 32%|███▏      | 15783/50000 [04:18<09:14, 61.67it/s]




 32%|███▏      | 15792/50000 [04:18<08:47, 64.91it/s]




 32%|███▏      | 15799/50000 [04:18<09:21, 60.87it/s]




 32%|███▏      | 15806/50000 [04:18<09:08, 62.38it/s]




 32%|███▏      | 15813/50000 [04:19<08:55, 63.83it/s]




 32%|███▏      | 15820/50000 [04:19<08:48, 64.70it/s]




 32%|███▏      | 15827/50000 [04:19<08:46, 64.93it/s]




 32%|███▏      | 15834/50000 [04:19<08:38, 65.95it/s]




 32%|███▏      | 15841/50000 [04:19<09:01, 63.11it/s]




 32%|███▏      | 15850/50000 [04:19<08:32, 66.68it/s]




 32%|███▏      | 15857/50000 [04:19<08:35, 66.25it/s]




 32%|███▏      | 15864/50000 [04:19<08:45, 64.95it/s]




 32%|███▏      | 15872/50000 [04:19<08:29, 66.99it/s]




 32%|███▏      | 15879/50000 [04:20<08:38, 65.81it/s]




 32%|███▏      | 15886/50000 [04:20<08:37, 65.93it/s]




 32%|███▏      | 15895/50000 [04:20<08:12, 69.25it/s]




 32%|███▏      | 15903/50000 [04:20<08:04, 70.31it/s]




 32%|███▏      | 15911/50000 [04:20<08:37, 65.83it/s]




 32%|███▏      | 15919/50000 [04:20<08:13, 69.12it/s]




 32%|███▏      | 15927/50000 [04:20<08:00, 70.96it/s]




 32%|███▏      | 15935/50000 [04:20<08:27, 67.13it/s]




 32%|███▏      | 15942/50000 [04:21<08:23, 67.62it/s]




 32%|███▏      | 15949/50000 [04:21<08:34, 66.25it/s]




 32%|███▏      | 15958/50000 [04:21<07:59, 70.97it/s]




 32%|███▏      | 15966/50000 [04:21<08:02, 70.60it/s]




 32%|███▏      | 15975/50000 [04:21<07:39, 74.04it/s]




 32%|███▏      | 15983/50000 [04:21<08:05, 70.04it/s]




 32%|███▏      | 15991/50000 [04:21<07:53, 71.83it/s]




 32%|███▏      | 15999/50000 [04:21<07:43, 73.34it/s]




 32%|███▏      | 16007/50000 [04:21<07:39, 74.02it/s]




 32%|███▏      | 16015/50000 [04:22<08:01, 70.58it/s]




 32%|███▏      | 16023/50000 [04:22<08:05, 69.96it/s]




 32%|███▏      | 16031/50000 [04:22<08:07, 69.71it/s]




 32%|███▏      | 16039/50000 [04:22<08:50, 64.05it/s]




 32%|███▏      | 16046/50000 [04:22<08:40, 65.22it/s]




 32%|███▏      | 16054/50000 [04:22<08:17, 68.22it/s]




 32%|███▏      | 16061/50000 [04:22<08:20, 67.80it/s]




 32%|███▏      | 16068/50000 [04:22<08:21, 67.70it/s]




 32%|███▏      | 16075/50000 [04:22<08:50, 63.94it/s]




 32%|███▏      | 16082/50000 [04:23<08:53, 63.55it/s]




 32%|███▏      | 16089/50000 [04:23<09:20, 60.50it/s]




 32%|███▏      | 16096/50000 [04:23<09:23, 60.19it/s]




 32%|███▏      | 16104/50000 [04:23<08:51, 63.82it/s]




 32%|███▏      | 16111/50000 [04:23<08:55, 63.31it/s]




 32%|███▏      | 16118/50000 [04:23<09:05, 62.11it/s]




 32%|███▏      | 16127/50000 [04:23<08:33, 65.97it/s]




 32%|███▏      | 16134/50000 [04:23<08:33, 65.92it/s]




 32%|███▏      | 16141/50000 [04:23<08:28, 66.57it/s]




 32%|███▏      | 16148/50000 [04:24<08:22, 67.35it/s]




 32%|███▏      | 16155/50000 [04:24<08:37, 65.41it/s]




 32%|███▏      | 16163/50000 [04:24<08:20, 67.58it/s]




 32%|███▏      | 16171/50000 [04:24<08:14, 68.39it/s]




 32%|███▏      | 16178/50000 [04:24<08:28, 66.57it/s]




 32%|███▏      | 16185/50000 [04:24<08:38, 65.16it/s]




 32%|███▏      | 16193/50000 [04:24<08:12, 68.69it/s]




 32%|███▏      | 16200/50000 [04:24<08:11, 68.73it/s]




 32%|███▏      | 16208/50000 [04:24<08:05, 69.57it/s]




 32%|███▏      | 16215/50000 [04:25<08:11, 68.73it/s]




 32%|███▏      | 16223/50000 [04:25<08:02, 69.94it/s]




 32%|███▏      | 16231/50000 [04:25<08:18, 67.76it/s]




 32%|███▏      | 16238/50000 [04:25<08:30, 66.14it/s]




 32%|███▏      | 16245/50000 [04:25<09:03, 62.12it/s]




 33%|███▎      | 16252/50000 [04:25<09:04, 61.96it/s]




 33%|███▎      | 16259/50000 [04:25<08:49, 63.69it/s]




 33%|███▎      | 16266/50000 [04:25<08:45, 64.25it/s]




 33%|███▎      | 16273/50000 [04:25<08:37, 65.19it/s]




 33%|███▎      | 16280/50000 [04:26<08:37, 65.13it/s]




 33%|███▎      | 16288/50000 [04:26<08:26, 66.62it/s]




 33%|███▎      | 16295/50000 [04:26<08:34, 65.56it/s]




 33%|███▎      | 16302/50000 [04:26<08:31, 65.94it/s]




 33%|███▎      | 16310/50000 [04:26<08:07, 69.12it/s]




 33%|███▎      | 16318/50000 [04:26<07:49, 71.72it/s]




 33%|███▎      | 16326/50000 [04:26<07:47, 72.08it/s]




 33%|███▎      | 16334/50000 [04:26<07:46, 72.13it/s]




 33%|███▎      | 16342/50000 [04:26<08:00, 70.00it/s]




 33%|███▎      | 16350/50000 [04:27<07:49, 71.61it/s]




 33%|███▎      | 16358/50000 [04:27<07:58, 70.29it/s]




 33%|███▎      | 16366/50000 [04:27<08:13, 68.16it/s]




 33%|███▎      | 16373/50000 [04:27<08:30, 65.86it/s]




 33%|███▎      | 16381/50000 [04:27<08:10, 68.52it/s]




 33%|███▎      | 16388/50000 [04:27<08:11, 68.41it/s]




 33%|███▎      | 16395/50000 [04:27<08:36, 65.10it/s]




 33%|███▎      | 16404/50000 [04:27<08:04, 69.40it/s]




 33%|███▎      | 16412/50000 [04:27<07:49, 71.55it/s]




 33%|███▎      | 16420/50000 [04:28<07:39, 73.13it/s]




 33%|███▎      | 16428/50000 [04:28<07:45, 72.09it/s]




 33%|███▎      | 16436/50000 [04:28<07:41, 72.68it/s]




 33%|███▎      | 16444/50000 [04:28<07:46, 71.92it/s]




 33%|███▎      | 16452/50000 [04:28<08:10, 68.34it/s]




 33%|███▎      | 16459/50000 [04:28<08:11, 68.28it/s]




 33%|███▎      | 16466/50000 [04:28<08:31, 65.56it/s]




 33%|███▎      | 16473/50000 [04:28<08:24, 66.50it/s]




 33%|███▎      | 16480/50000 [04:28<08:17, 67.37it/s]




 33%|███▎      | 16488/50000 [04:29<07:57, 70.21it/s]




 33%|███▎      | 16496/50000 [04:29<07:59, 69.89it/s]




 33%|███▎      | 16504/50000 [04:29<07:44, 72.11it/s]




 33%|███▎      | 16512/50000 [04:29<07:48, 71.48it/s]




 33%|███▎      | 16520/50000 [04:29<08:28, 65.88it/s]




 33%|███▎      | 16527/50000 [04:29<09:24, 59.29it/s]




 33%|███▎      | 16534/50000 [04:29<09:46, 57.02it/s]




 33%|███▎      | 16540/50000 [04:29<10:02, 55.52it/s]




 33%|███▎      | 16546/50000 [04:30<11:07, 50.15it/s]




 33%|███▎      | 16552/50000 [04:30<10:38, 52.37it/s]




 33%|███▎      | 16558/50000 [04:30<10:48, 51.55it/s]




 33%|███▎      | 16564/50000 [04:30<10:50, 51.38it/s]




 33%|███▎      | 16570/50000 [04:30<11:18, 49.24it/s]




 33%|███▎      | 16576/50000 [04:30<11:13, 49.63it/s]




 33%|███▎      | 16582/50000 [04:30<11:17, 49.29it/s]




 33%|███▎      | 16588/50000 [04:30<10:59, 50.67it/s]




 33%|███▎      | 16594/50000 [04:30<10:33, 52.76it/s]




 33%|███▎      | 16601/50000 [04:31<09:48, 56.73it/s]




 33%|███▎      | 16609/50000 [04:31<09:00, 61.76it/s]




 33%|███▎      | 16616/50000 [04:31<08:54, 62.50it/s]




 33%|███▎      | 16623/50000 [04:31<09:00, 61.81it/s]




 33%|███▎      | 16631/50000 [04:31<08:29, 65.50it/s]




 33%|███▎      | 16638/50000 [04:31<08:39, 64.18it/s]




 33%|███▎      | 16646/50000 [04:31<08:27, 65.73it/s]




 33%|███▎      | 16653/50000 [04:31<08:33, 64.94it/s]




 33%|███▎      | 16660/50000 [04:31<08:31, 65.22it/s]




 33%|███▎      | 16667/50000 [04:32<08:51, 62.71it/s]




 33%|███▎      | 16674/50000 [04:32<09:05, 61.12it/s]




 33%|███▎      | 16681/50000 [04:32<09:26, 58.82it/s]




 33%|███▎      | 16687/50000 [04:32<10:15, 54.11it/s]




 33%|███▎      | 16693/50000 [04:32<10:20, 53.70it/s]




 33%|███▎      | 16699/50000 [04:32<10:49, 51.23it/s]




 33%|███▎      | 16705/50000 [04:32<10:44, 51.69it/s]




 33%|███▎      | 16711/50000 [04:32<10:40, 52.01it/s]




 33%|███▎      | 16718/50000 [04:33<10:09, 54.58it/s]




 33%|███▎      | 16726/50000 [04:33<09:17, 59.70it/s]




 33%|███▎      | 16733/50000 [04:33<08:53, 62.33it/s]




 33%|███▎      | 16742/50000 [04:33<08:21, 66.37it/s]




 33%|███▎      | 16749/50000 [04:33<08:56, 61.93it/s]




 34%|███▎      | 16757/50000 [04:33<08:25, 65.82it/s]




 34%|███▎      | 16764/50000 [04:33<08:24, 65.94it/s]




 34%|███▎      | 16771/50000 [04:33<08:20, 66.40it/s]




 34%|███▎      | 16778/50000 [04:33<08:19, 66.53it/s]




 34%|███▎      | 16785/50000 [04:34<08:52, 62.36it/s]




 34%|███▎      | 16792/50000 [04:34<08:38, 63.99it/s]




 34%|███▎      | 16799/50000 [04:34<08:27, 65.36it/s]




 34%|███▎      | 16807/50000 [04:34<08:03, 68.67it/s]




 34%|███▎      | 16814/50000 [04:34<08:20, 66.28it/s]




 34%|███▎      | 16821/50000 [04:34<08:50, 62.54it/s]




 34%|███▎      | 16829/50000 [04:34<08:33, 64.54it/s]




 34%|███▎      | 16836/50000 [04:34<08:51, 62.43it/s]




 34%|███▎      | 16844/50000 [04:34<08:27, 65.35it/s]




 34%|███▎      | 16851/50000 [04:35<08:28, 65.15it/s]




 34%|███▎      | 16859/50000 [04:35<08:13, 67.12it/s]




 34%|███▎      | 16866/50000 [04:35<08:20, 66.19it/s]




 34%|███▎      | 16873/50000 [04:35<08:15, 66.91it/s]




 34%|███▍      | 16880/50000 [04:35<08:23, 65.76it/s]




 34%|███▍      | 16887/50000 [04:35<08:23, 65.71it/s]




 34%|███▍      | 16896/50000 [04:35<07:53, 69.88it/s]




 34%|███▍      | 16904/50000 [04:35<07:58, 69.12it/s]




 34%|███▍      | 16912/50000 [04:35<07:56, 69.49it/s]




 34%|███▍      | 16920/50000 [04:36<07:55, 69.57it/s]




 34%|███▍      | 16927/50000 [04:36<08:01, 68.73it/s]




 34%|███▍      | 16934/50000 [04:36<08:03, 68.35it/s]




 34%|███▍      | 16942/50000 [04:36<07:58, 69.03it/s]




 34%|███▍      | 16949/50000 [04:36<08:09, 67.57it/s]




 34%|███▍      | 16956/50000 [04:36<08:04, 68.14it/s]




 34%|███▍      | 16963/50000 [04:36<08:24, 65.47it/s]




 34%|███▍      | 16970/50000 [04:36<08:39, 63.55it/s]




 34%|███▍      | 16977/50000 [04:36<08:49, 62.35it/s]




 34%|███▍      | 16984/50000 [04:37<08:43, 63.12it/s]




 34%|███▍      | 16991/50000 [04:37<08:41, 63.32it/s]




 34%|███▍      | 16999/50000 [04:37<08:20, 65.93it/s]




 34%|███▍      | 17006/50000 [04:37<08:14, 66.77it/s]




 34%|███▍      | 17013/50000 [04:37<08:33, 64.22it/s]




 34%|███▍      | 17020/50000 [04:37<08:30, 64.63it/s]




 34%|███▍      | 17027/50000 [04:37<08:19, 66.01it/s]




 34%|███▍      | 17034/50000 [04:37<08:36, 63.82it/s]




 34%|███▍      | 17041/50000 [04:37<08:32, 64.34it/s]




 34%|███▍      | 17048/50000 [04:38<08:24, 65.32it/s]




 34%|███▍      | 17055/50000 [04:38<08:19, 65.95it/s]




 34%|███▍      | 17062/50000 [04:38<08:19, 65.92it/s]




 34%|███▍      | 17069/50000 [04:38<08:14, 66.53it/s]




 34%|███▍      | 17077/50000 [04:38<07:57, 68.95it/s]




 34%|███▍      | 17085/50000 [04:38<07:54, 69.33it/s]




 34%|███▍      | 17093/50000 [04:38<07:45, 70.67it/s]




 34%|███▍      | 17101/50000 [04:38<07:56, 69.03it/s]




 34%|███▍      | 17108/50000 [04:38<08:05, 67.76it/s]




 34%|███▍      | 17116/50000 [04:38<07:56, 69.05it/s]




 34%|███▍      | 17123/50000 [04:39<07:56, 68.98it/s]




 34%|███▍      | 17130/50000 [04:39<08:16, 66.20it/s]




 34%|███▍      | 17138/50000 [04:39<07:56, 68.96it/s]




 34%|███▍      | 17146/50000 [04:39<07:48, 70.10it/s]




 34%|███▍      | 17154/50000 [04:39<07:45, 70.55it/s]




 34%|███▍      | 17162/50000 [04:39<07:40, 71.25it/s]




 34%|███▍      | 17170/50000 [04:39<07:55, 69.02it/s]




 34%|███▍      | 17178/50000 [04:39<07:41, 71.08it/s]




 34%|███▍      | 17186/50000 [04:39<07:49, 69.94it/s]




 34%|███▍      | 17194/50000 [04:40<08:01, 68.07it/s]




 34%|███▍      | 17201/50000 [04:40<07:59, 68.39it/s]




 34%|███▍      | 17208/50000 [04:40<07:58, 68.51it/s]




 34%|███▍      | 17216/50000 [04:40<07:53, 69.24it/s]




 34%|███▍      | 17223/50000 [04:40<07:58, 68.50it/s]




 34%|███▍      | 17230/50000 [04:40<08:14, 66.26it/s]




 34%|███▍      | 17237/50000 [04:40<08:14, 66.25it/s]




 34%|███▍      | 17245/50000 [04:40<07:58, 68.47it/s]




 35%|███▍      | 17253/50000 [04:40<07:50, 69.56it/s]




 35%|███▍      | 17260/50000 [04:41<08:00, 68.13it/s]




 35%|███▍      | 17267/50000 [04:41<08:06, 67.35it/s]




 35%|███▍      | 17275/50000 [04:41<07:59, 68.28it/s]




 35%|███▍      | 17282/50000 [04:41<08:20, 65.38it/s]




 35%|███▍      | 17289/50000 [04:41<08:19, 65.51it/s]




 35%|███▍      | 17296/50000 [04:41<08:34, 63.54it/s]




 35%|███▍      | 17303/50000 [04:41<08:31, 63.87it/s]




 35%|███▍      | 17310/50000 [04:41<08:45, 62.16it/s]




 35%|███▍      | 17317/50000 [04:41<09:02, 60.22it/s]




 35%|███▍      | 17324/50000 [04:42<09:01, 60.31it/s]




 35%|███▍      | 17331/50000 [04:42<09:20, 58.26it/s]




 35%|███▍      | 17340/50000 [04:42<08:22, 65.05it/s]




 35%|███▍      | 17347/50000 [04:42<08:24, 64.67it/s]




 35%|███▍      | 17355/50000 [04:42<08:02, 67.62it/s]




 35%|███▍      | 17362/50000 [04:42<08:27, 64.33it/s]




 35%|███▍      | 17369/50000 [04:42<08:31, 63.82it/s]




 35%|███▍      | 17376/50000 [04:42<08:29, 63.99it/s]




 35%|███▍      | 17383/50000 [04:43<08:42, 62.40it/s]




 35%|███▍      | 17390/50000 [04:43<08:46, 61.90it/s]




 35%|███▍      | 17397/50000 [04:43<08:41, 62.54it/s]




 35%|███▍      | 17404/50000 [04:43<08:42, 62.33it/s]




 35%|███▍      | 17412/50000 [04:43<08:11, 66.30it/s]




 35%|███▍      | 17419/50000 [04:43<08:18, 65.35it/s]




 35%|███▍      | 17426/50000 [04:43<08:15, 65.79it/s]




 35%|███▍      | 17433/50000 [04:43<08:12, 66.10it/s]




 35%|███▍      | 17440/50000 [04:43<08:26, 64.32it/s]




 35%|███▍      | 17447/50000 [04:44<08:32, 63.47it/s]




 35%|███▍      | 17454/50000 [04:44<08:19, 65.16it/s]




 35%|███▍      | 17461/50000 [04:44<08:14, 65.84it/s]




 35%|███▍      | 17469/50000 [04:44<08:06, 66.88it/s]




 35%|███▍      | 17477/50000 [04:44<07:49, 69.30it/s]




 35%|███▍      | 17485/50000 [04:44<07:45, 69.79it/s]




 35%|███▍      | 17493/50000 [04:44<08:03, 67.25it/s]




 35%|███▌      | 17500/50000 [04:44<08:32, 63.39it/s]




 35%|███▌      | 17507/50000 [04:44<08:46, 61.75it/s]




 35%|███▌      | 17514/50000 [04:45<08:58, 60.34it/s]




 35%|███▌      | 17521/50000 [04:45<08:56, 60.55it/s]




 35%|███▌      | 17528/50000 [04:45<08:48, 61.49it/s]




 35%|███▌      | 17536/50000 [04:45<08:18, 65.15it/s]




 35%|███▌      | 17543/50000 [04:45<08:32, 63.34it/s]




 35%|███▌      | 17550/50000 [04:45<09:15, 58.41it/s]




 35%|███▌      | 17556/50000 [04:45<09:26, 57.24it/s]




 35%|███▌      | 17562/50000 [04:45<09:20, 57.83it/s]




 35%|███▌      | 17568/50000 [04:45<09:43, 55.59it/s]




 35%|███▌      | 17574/50000 [04:46<09:55, 54.42it/s]




 35%|███▌      | 17580/50000 [04:46<10:07, 53.34it/s]




 35%|███▌      | 17586/50000 [04:46<10:27, 51.66it/s]




 35%|███▌      | 17592/50000 [04:46<10:40, 50.61it/s]




 35%|███▌      | 17598/50000 [04:46<11:39, 46.35it/s]




 35%|███▌      | 17603/50000 [04:46<12:34, 42.92it/s]




 35%|███▌      | 17608/50000 [04:46<12:11, 44.26it/s]




 35%|███▌      | 17615/50000 [04:46<11:08, 48.44it/s]




 35%|███▌      | 17623/50000 [04:47<09:51, 54.75it/s]




 35%|███▌      | 17630/50000 [04:47<09:21, 57.60it/s]




 35%|███▌      | 17637/50000 [04:47<08:55, 60.41it/s]




 35%|███▌      | 17644/50000 [04:47<08:42, 61.88it/s]




 35%|███▌      | 17652/50000 [04:47<08:09, 66.02it/s]




 35%|███▌      | 17660/50000 [04:47<07:45, 69.54it/s]




 35%|███▌      | 17668/50000 [04:47<07:48, 69.06it/s]




 35%|███▌      | 17676/50000 [04:47<07:46, 69.27it/s]




 35%|███▌      | 17685/50000 [04:47<07:33, 71.28it/s]




 35%|███▌      | 17693/50000 [04:48<08:06, 66.46it/s]




 35%|███▌      | 17701/50000 [04:48<07:42, 69.86it/s]




 35%|███▌      | 17709/50000 [04:48<08:12, 65.63it/s]




 35%|███▌      | 17716/50000 [04:48<08:38, 62.21it/s]




 35%|███▌      | 17723/50000 [04:48<09:09, 58.74it/s]




 35%|███▌      | 17730/50000 [04:48<09:39, 55.73it/s]




 35%|███▌      | 17736/50000 [04:48<09:42, 55.42it/s]




 35%|███▌      | 17743/50000 [04:48<09:14, 58.12it/s]




 36%|███▌      | 17750/50000 [04:49<08:50, 60.80it/s]




 36%|███▌      | 17757/50000 [04:49<08:31, 63.00it/s]




 36%|███▌      | 17765/50000 [04:49<08:01, 67.00it/s]




 36%|███▌      | 17772/50000 [04:49<08:06, 66.20it/s]




 36%|███▌      | 17779/50000 [04:49<08:21, 64.20it/s]




 36%|███▌      | 17787/50000 [04:49<08:03, 66.67it/s]




 36%|███▌      | 17794/50000 [04:49<08:05, 66.27it/s]




 36%|███▌      | 17802/50000 [04:49<07:50, 68.48it/s]




 36%|███▌      | 17809/50000 [04:49<07:50, 68.38it/s]




 36%|███▌      | 17817/50000 [04:49<07:47, 68.84it/s]




 36%|███▌      | 17824/50000 [04:50<08:10, 65.62it/s]




 36%|███▌      | 17831/50000 [04:50<08:10, 65.61it/s]




 36%|███▌      | 17839/50000 [04:50<07:56, 67.48it/s]




 36%|███▌      | 17847/50000 [04:50<07:39, 69.93it/s]




 36%|███▌      | 17855/50000 [04:50<07:24, 72.33it/s]




 36%|███▌      | 17863/50000 [04:50<07:34, 70.78it/s]




 36%|███▌      | 17871/50000 [04:50<07:29, 71.41it/s]




 36%|███▌      | 17879/50000 [04:50<07:34, 70.72it/s]




 36%|███▌      | 17887/50000 [04:51<07:54, 67.75it/s]




 36%|███▌      | 17895/50000 [04:51<07:42, 69.49it/s]




 36%|███▌      | 17902/50000 [04:51<07:47, 68.67it/s]




 36%|███▌      | 17909/50000 [04:51<07:56, 67.33it/s]




 36%|███▌      | 17916/50000 [04:51<08:09, 65.49it/s]




 36%|███▌      | 17923/50000 [04:51<08:06, 65.94it/s]




 36%|███▌      | 17930/50000 [04:51<08:09, 65.47it/s]




 36%|███▌      | 17937/50000 [04:51<08:36, 62.09it/s]




 36%|███▌      | 17946/50000 [04:51<07:54, 67.53it/s]




 36%|███▌      | 17953/50000 [04:51<07:57, 67.13it/s]




 36%|███▌      | 17961/50000 [04:52<07:45, 68.77it/s]




 36%|███▌      | 17968/50000 [04:52<07:47, 68.58it/s]




 36%|███▌      | 17975/50000 [04:52<07:58, 66.88it/s]




 36%|███▌      | 17983/50000 [04:52<07:48, 68.41it/s]




 36%|███▌      | 17991/50000 [04:52<07:35, 70.25it/s]




 36%|███▌      | 17999/50000 [04:52<07:38, 69.74it/s]




 36%|███▌      | 18007/50000 [04:52<07:44, 68.84it/s]




 36%|███▌      | 18015/50000 [04:52<07:46, 68.58it/s]




 36%|███▌      | 18022/50000 [04:52<07:46, 68.53it/s]




 36%|███▌      | 18029/50000 [04:53<07:52, 67.62it/s]




 36%|███▌      | 18036/50000 [04:53<07:53, 67.46it/s]




 36%|███▌      | 18043/50000 [04:53<07:57, 66.89it/s]




 36%|███▌      | 18050/50000 [04:53<08:01, 66.40it/s]




 36%|███▌      | 18058/50000 [04:53<07:51, 67.71it/s]




 36%|███▌      | 18065/50000 [04:53<07:52, 67.64it/s]




 36%|███▌      | 18072/50000 [04:53<07:56, 67.02it/s]




 36%|███▌      | 18079/50000 [04:53<07:51, 67.74it/s]




 36%|███▌      | 18086/50000 [04:53<08:36, 61.77it/s]




 36%|███▌      | 18093/50000 [04:54<08:58, 59.21it/s]




 36%|███▌      | 18100/50000 [04:54<09:13, 57.68it/s]




 36%|███▌      | 18107/50000 [04:54<09:01, 58.94it/s]




 36%|███▌      | 18113/50000 [04:54<09:04, 58.52it/s]




 36%|███▌      | 18119/50000 [04:54<09:37, 55.21it/s]




 36%|███▋      | 18125/50000 [04:54<09:31, 55.82it/s]




 36%|███▋      | 18131/50000 [04:54<09:21, 56.73it/s]




 36%|███▋      | 18137/50000 [04:54<09:45, 54.43it/s]




 36%|███▋      | 18143/50000 [04:55<09:49, 54.07it/s]




 36%|███▋      | 18149/50000 [04:55<09:48, 54.11it/s]




 36%|███▋      | 18156/50000 [04:55<09:14, 57.39it/s]




 36%|███▋      | 18163/50000 [04:55<08:48, 60.24it/s]




 36%|███▋      | 18170/50000 [04:55<09:10, 57.79it/s]




 36%|███▋      | 18177/50000 [04:55<08:49, 60.08it/s]




 36%|███▋      | 18184/50000 [04:55<08:42, 60.83it/s]




 36%|███▋      | 18191/50000 [04:55<08:45, 60.58it/s]




 36%|███▋      | 18198/50000 [04:55<08:39, 61.19it/s]




 36%|███▋      | 18205/50000 [04:56<08:35, 61.62it/s]




 36%|███▋      | 18212/50000 [04:56<08:48, 60.18it/s]




 36%|███▋      | 18219/50000 [04:56<09:02, 58.62it/s]




 36%|███▋      | 18226/50000 [04:56<08:50, 59.93it/s]




 36%|███▋      | 18233/50000 [04:56<09:05, 58.23it/s]




 36%|███▋      | 18241/50000 [04:56<08:39, 61.12it/s]




 36%|███▋      | 18248/50000 [04:56<09:33, 55.32it/s]




 37%|███▋      | 18254/50000 [04:56<09:35, 55.13it/s]




 37%|███▋      | 18260/50000 [04:57<09:52, 53.53it/s]




 37%|███▋      | 18266/50000 [04:57<10:55, 48.41it/s]




 37%|███▋      | 18271/50000 [04:57<11:01, 47.93it/s]




 37%|███▋      | 18277/50000 [04:57<11:27, 46.17it/s]




 37%|███▋      | 18283/50000 [04:57<10:50, 48.78it/s]




 37%|███▋      | 18288/50000 [04:57<10:54, 48.47it/s]




 37%|███▋      | 18294/50000 [04:57<10:42, 49.32it/s]




 37%|███▋      | 18299/50000 [04:57<11:22, 46.43it/s]




 37%|███▋      | 18305/50000 [04:57<10:58, 48.11it/s]




 37%|███▋      | 18311/50000 [04:58<11:03, 47.74it/s]




 37%|███▋      | 18316/50000 [04:58<12:40, 41.67it/s]




 37%|███▋      | 18323/50000 [04:58<11:21, 46.51it/s]




 37%|███▋      | 18330/50000 [04:58<10:21, 50.96it/s]




 37%|███▋      | 18336/50000 [04:58<10:13, 51.62it/s]




 37%|███▋      | 18344/50000 [04:58<09:19, 56.58it/s]




 37%|███▋      | 18351/50000 [04:58<08:57, 58.86it/s]




 37%|███▋      | 18359/50000 [04:58<08:28, 62.19it/s]




 37%|███▋      | 18366/50000 [04:59<08:41, 60.71it/s]




 37%|███▋      | 18374/50000 [04:59<08:19, 63.32it/s]




 37%|███▋      | 18381/50000 [04:59<08:20, 63.12it/s]




 37%|███▋      | 18388/50000 [04:59<08:43, 60.38it/s]




 37%|███▋      | 18395/50000 [04:59<08:29, 62.02it/s]




 37%|███▋      | 18402/50000 [04:59<08:13, 64.00it/s]




 37%|███▋      | 18409/50000 [04:59<08:20, 63.16it/s]




 37%|███▋      | 18416/50000 [04:59<08:07, 64.76it/s]




 37%|███▋      | 18423/50000 [04:59<08:09, 64.47it/s]




 37%|███▋      | 18430/50000 [05:00<09:01, 58.34it/s]




 37%|███▋      | 18436/50000 [05:00<09:07, 57.69it/s]




 37%|███▋      | 18443/50000 [05:00<08:48, 59.70it/s]




 37%|███▋      | 18450/50000 [05:00<08:32, 61.52it/s]




 37%|███▋      | 18457/50000 [05:00<08:28, 62.02it/s]




 37%|███▋      | 18465/50000 [05:00<08:06, 64.78it/s]




 37%|███▋      | 18472/50000 [05:00<08:27, 62.09it/s]




 37%|███▋      | 18480/50000 [05:00<08:06, 64.77it/s]




 37%|███▋      | 18487/50000 [05:00<08:13, 63.86it/s]




 37%|███▋      | 18494/50000 [05:01<08:20, 62.90it/s]




 37%|███▋      | 18501/50000 [05:01<08:30, 61.67it/s]




 37%|███▋      | 18508/50000 [05:01<08:28, 61.96it/s]




 37%|███▋      | 18515/50000 [05:01<08:34, 61.20it/s]




 37%|███▋      | 18522/50000 [05:01<09:25, 55.62it/s]




 37%|███▋      | 18528/50000 [05:01<10:23, 50.47it/s]




 37%|███▋      | 18534/50000 [05:01<11:02, 47.50it/s]




 37%|███▋      | 18539/50000 [05:01<10:56, 47.91it/s]




 37%|███▋      | 18544/50000 [05:02<11:43, 44.72it/s]




 37%|███▋      | 18549/50000 [05:02<11:46, 44.50it/s]




 37%|███▋      | 18554/50000 [05:02<11:30, 45.55it/s]




 37%|███▋      | 18559/50000 [05:02<11:54, 44.00it/s]




 37%|███▋      | 18564/50000 [05:02<13:05, 40.04it/s]




 37%|███▋      | 18569/50000 [05:02<12:59, 40.30it/s]




 37%|███▋      | 18574/50000 [05:02<12:43, 41.18it/s]




 37%|███▋      | 18579/50000 [05:02<12:06, 43.23it/s]




 37%|███▋      | 18584/50000 [05:03<12:27, 42.03it/s]




 37%|███▋      | 18589/50000 [05:03<12:19, 42.49it/s]




 37%|███▋      | 18594/50000 [05:03<12:13, 42.81it/s]




 37%|███▋      | 18601/50000 [05:03<10:59, 47.60it/s]




 37%|███▋      | 18609/50000 [05:03<09:51, 53.03it/s]




 37%|███▋      | 18616/50000 [05:03<09:19, 56.13it/s]




 37%|███▋      | 18624/50000 [05:03<08:46, 59.62it/s]




 37%|███▋      | 18631/50000 [05:03<08:52, 58.90it/s]




 37%|███▋      | 18638/50000 [05:03<08:50, 59.07it/s]




 37%|███▋      | 18645/50000 [05:04<09:25, 55.41it/s]




 37%|███▋      | 18651/50000 [05:04<09:48, 53.29it/s]




 37%|███▋      | 18657/50000 [05:04<10:00, 52.23it/s]




 37%|███▋      | 18663/50000 [05:04<10:09, 51.40it/s]




 37%|███▋      | 18669/50000 [05:04<09:46, 53.45it/s]




 37%|███▋      | 18675/50000 [05:04<10:48, 48.31it/s]




 37%|███▋      | 18680/50000 [05:04<11:18, 46.14it/s]




 37%|███▋      | 18685/50000 [05:04<11:36, 44.98it/s]




 37%|███▋      | 18690/50000 [05:05<11:44, 44.44it/s]




 37%|███▋      | 18695/50000 [05:05<11:42, 44.53it/s]




 37%|███▋      | 18700/50000 [05:05<11:21, 45.95it/s]




 37%|███▋      | 18706/50000 [05:05<10:37, 49.09it/s]




 37%|███▋      | 18713/50000 [05:05<09:42, 53.71it/s]




 37%|███▋      | 18719/50000 [05:05<09:36, 54.22it/s]




 37%|███▋      | 18726/50000 [05:05<09:01, 57.76it/s]




 37%|███▋      | 18734/50000 [05:05<08:20, 62.46it/s]




 37%|███▋      | 18741/50000 [05:05<08:18, 62.69it/s]




 37%|███▋      | 18748/50000 [05:06<08:11, 63.59it/s]




 38%|███▊      | 18755/50000 [05:06<08:00, 65.07it/s]




 38%|███▊      | 18762/50000 [05:06<08:14, 63.20it/s]




 38%|███▊      | 18769/50000 [05:06<08:24, 61.95it/s]




 38%|███▊      | 18777/50000 [05:06<08:07, 64.10it/s]




 38%|███▊      | 18785/50000 [05:06<07:53, 65.86it/s]




 38%|███▊      | 18792/50000 [05:06<07:50, 66.34it/s]




 38%|███▊      | 18799/50000 [05:06<08:31, 60.95it/s]




 38%|███▊      | 18806/50000 [05:06<08:28, 61.29it/s]




 38%|███▊      | 18813/50000 [05:07<08:33, 60.74it/s]




 38%|███▊      | 18820/50000 [05:07<08:43, 59.51it/s]




 38%|███▊      | 18828/50000 [05:07<08:07, 63.89it/s]




 38%|███▊      | 18835/50000 [05:07<08:34, 60.58it/s]




 38%|███▊      | 18843/50000 [05:07<08:05, 64.12it/s]




 38%|███▊      | 18850/50000 [05:07<08:04, 64.28it/s]




 38%|███▊      | 18857/50000 [05:07<08:05, 64.14it/s]




 38%|███▊      | 18864/50000 [05:07<08:12, 63.26it/s]




 38%|███▊      | 18871/50000 [05:07<08:09, 63.59it/s]




 38%|███▊      | 18878/50000 [05:08<08:11, 63.31it/s]




 38%|███▊      | 18885/50000 [05:08<07:58, 65.05it/s]




 38%|███▊      | 18892/50000 [05:08<08:08, 63.62it/s]




 38%|███▊      | 18899/50000 [05:08<08:12, 63.16it/s]




 38%|███▊      | 18906/50000 [05:08<08:20, 62.17it/s]




 38%|███▊      | 18913/50000 [05:08<08:13, 62.99it/s]




 38%|███▊      | 18921/50000 [05:08<08:07, 63.79it/s]




 38%|███▊      | 18928/50000 [05:08<08:24, 61.61it/s]




 38%|███▊      | 18935/50000 [05:09<08:44, 59.26it/s]




 38%|███▊      | 18941/50000 [05:09<09:00, 57.51it/s]




 38%|███▊      | 18947/50000 [05:09<09:01, 57.35it/s]




 38%|███▊      | 18955/50000 [05:09<08:22, 61.84it/s]




 38%|███▊      | 18962/50000 [05:09<08:06, 63.78it/s]




 38%|███▊      | 18969/50000 [05:09<08:05, 63.96it/s]




 38%|███▊      | 18977/50000 [05:09<07:51, 65.76it/s]




 38%|███▊      | 18984/50000 [05:09<07:52, 65.71it/s]




 38%|███▊      | 18991/50000 [05:09<08:12, 63.02it/s]




 38%|███▊      | 18999/50000 [05:09<07:51, 65.70it/s]




 38%|███▊      | 19006/50000 [05:10<07:46, 66.41it/s]




 38%|███▊      | 19013/50000 [05:10<07:46, 66.44it/s]




 38%|███▊      | 19020/50000 [05:10<08:00, 64.45it/s]




 38%|███▊      | 19027/50000 [05:10<08:00, 64.41it/s]




 38%|███▊      | 19034/50000 [05:10<08:15, 62.53it/s]




 38%|███▊      | 19041/50000 [05:10<08:50, 58.36it/s]




 38%|███▊      | 19049/50000 [05:10<08:19, 61.96it/s]




 38%|███▊      | 19056/50000 [05:10<08:09, 63.23it/s]




 38%|███▊      | 19064/50000 [05:11<07:43, 66.68it/s]




 38%|███▊      | 19071/50000 [05:11<08:02, 64.08it/s]




 38%|███▊      | 19079/50000 [05:11<07:42, 66.83it/s]




 38%|███▊      | 19086/50000 [05:11<07:39, 67.22it/s]




 38%|███▊      | 19095/50000 [05:11<07:18, 70.40it/s]




 38%|███▊      | 19103/50000 [05:11<07:15, 70.95it/s]




 38%|███▊      | 19111/50000 [05:11<07:06, 72.50it/s]




 38%|███▊      | 19119/50000 [05:11<07:14, 71.08it/s]




 38%|███▊      | 19127/50000 [05:11<08:00, 64.23it/s]




 38%|███▊      | 19134/50000 [05:12<08:23, 61.25it/s]




 38%|███▊      | 19142/50000 [05:12<08:00, 64.18it/s]




 38%|███▊      | 19149/50000 [05:12<07:56, 64.78it/s]




 38%|███▊      | 19156/50000 [05:12<08:12, 62.59it/s]




 38%|███▊      | 19163/50000 [05:12<08:13, 62.44it/s]




 38%|███▊      | 19170/50000 [05:12<08:34, 59.95it/s]




 38%|███▊      | 19177/50000 [05:12<08:40, 59.20it/s]




 38%|███▊      | 19183/50000 [05:12<08:59, 57.11it/s]




 38%|███▊      | 19190/50000 [05:12<08:35, 59.72it/s]




 38%|███▊      | 19197/50000 [05:13<08:17, 61.86it/s]




 38%|███▊      | 19204/50000 [05:13<08:44, 58.67it/s]




 38%|███▊      | 19210/50000 [05:13<09:23, 54.60it/s]




 38%|███▊      | 19217/50000 [05:13<08:56, 57.35it/s]




 38%|███▊      | 19224/50000 [05:13<08:28, 60.52it/s]




 38%|███▊      | 19231/50000 [05:13<08:35, 59.75it/s]




 38%|███▊      | 19238/50000 [05:13<09:45, 52.58it/s]




 38%|███▊      | 19244/50000 [05:13<10:05, 50.84it/s]




 38%|███▊      | 19250/50000 [05:14<10:37, 48.25it/s]




 39%|███▊      | 19255/50000 [05:14<10:39, 48.10it/s]




 39%|███▊      | 19262/50000 [05:14<09:50, 52.05it/s]




 39%|███▊      | 19268/50000 [05:14<10:15, 49.92it/s]




 39%|███▊      | 19274/50000 [05:14<10:48, 47.39it/s]




 39%|███▊      | 19280/50000 [05:14<10:16, 49.86it/s]




 39%|███▊      | 19286/50000 [05:14<10:05, 50.70it/s]




 39%|███▊      | 19292/50000 [05:14<09:56, 51.50it/s]




 39%|███▊      | 19299/50000 [05:15<09:26, 54.16it/s]




 39%|███▊      | 19307/50000 [05:15<08:41, 58.84it/s]




 39%|███▊      | 19314/50000 [05:15<09:09, 55.79it/s]




 39%|███▊      | 19320/50000 [05:15<09:22, 54.58it/s]




 39%|███▊      | 19326/50000 [05:15<10:12, 50.05it/s]




 39%|███▊      | 19332/50000 [05:15<10:28, 48.79it/s]




 39%|███▊      | 19338/50000 [05:15<09:54, 51.59it/s]




 39%|███▊      | 19344/50000 [05:15<10:05, 50.63it/s]




 39%|███▊      | 19351/50000 [05:16<09:29, 53.85it/s]




 39%|███▊      | 19357/50000 [05:16<09:15, 55.14it/s]




 39%|███▊      | 19363/50000 [05:16<09:03, 56.40it/s]




 39%|███▊      | 19370/50000 [05:16<08:44, 58.43it/s]




 39%|███▉      | 19377/50000 [05:16<08:25, 60.57it/s]




 39%|███▉      | 19384/50000 [05:16<08:40, 58.80it/s]




 39%|███▉      | 19390/50000 [05:16<08:57, 56.93it/s]




 39%|███▉      | 19398/50000 [05:16<08:25, 60.52it/s]




 39%|███▉      | 19405/50000 [05:16<08:12, 62.13it/s]




 39%|███▉      | 19412/50000 [05:17<08:05, 62.96it/s]




 39%|███▉      | 19419/50000 [05:17<09:04, 56.15it/s]




 39%|███▉      | 19427/50000 [05:17<08:25, 60.52it/s]




 39%|███▉      | 19434/50000 [05:17<08:42, 58.55it/s]




 39%|███▉      | 19441/50000 [05:17<09:08, 55.74it/s]




 39%|███▉      | 19447/50000 [05:17<09:34, 53.21it/s]




 39%|███▉      | 19453/50000 [05:17<10:10, 50.04it/s]




 39%|███▉      | 19459/50000 [05:17<09:54, 51.35it/s]




 39%|███▉      | 19465/50000 [05:18<11:19, 44.92it/s]




 39%|███▉      | 19470/50000 [05:18<11:21, 44.81it/s]




 39%|███▉      | 19475/50000 [05:18<11:15, 45.22it/s]




 39%|███▉      | 19480/50000 [05:18<11:29, 44.25it/s]




 39%|███▉      | 19486/50000 [05:18<10:53, 46.72it/s]




 39%|███▉      | 19492/50000 [05:18<10:19, 49.21it/s]




 39%|███▉      | 19498/50000 [05:18<10:06, 50.29it/s]




 39%|███▉      | 19504/50000 [05:18<09:52, 51.47it/s]




 39%|███▉      | 19510/50000 [05:18<10:22, 49.00it/s]




 39%|███▉      | 19515/50000 [05:19<10:24, 48.83it/s]




 39%|███▉      | 19522/50000 [05:19<09:33, 53.12it/s]




 39%|███▉      | 19529/50000 [05:19<09:29, 53.50it/s]




 39%|███▉      | 19535/50000 [05:19<09:22, 54.14it/s]




 39%|███▉      | 19542/50000 [05:19<08:55, 56.86it/s]




 39%|███▉      | 19548/50000 [05:19<09:20, 54.36it/s]




 39%|███▉      | 19555/50000 [05:19<09:21, 54.25it/s]




 39%|███▉      | 19561/50000 [05:19<09:06, 55.74it/s]




 39%|███▉      | 19567/50000 [05:20<09:02, 56.12it/s]




 39%|███▉      | 19574/50000 [05:20<08:41, 58.37it/s]




 39%|███▉      | 19580/50000 [05:20<09:05, 55.78it/s]




 39%|███▉      | 19586/50000 [05:20<09:02, 56.11it/s]




 39%|███▉      | 19594/50000 [05:20<08:17, 61.10it/s]




 39%|███▉      | 19601/50000 [05:20<08:52, 57.13it/s]




 39%|███▉      | 19607/50000 [05:20<09:13, 54.91it/s]




 39%|███▉      | 19613/50000 [05:20<10:07, 49.99it/s]




 39%|███▉      | 19619/50000 [05:20<10:20, 48.93it/s]




 39%|███▉      | 19625/50000 [05:21<10:58, 46.11it/s]




 39%|███▉      | 19632/50000 [05:21<10:08, 49.87it/s]




 39%|███▉      | 19638/50000 [05:21<10:01, 50.45it/s]




 39%|███▉      | 19644/50000 [05:21<09:46, 51.79it/s]




 39%|███▉      | 19650/50000 [05:21<09:26, 53.61it/s]




 39%|███▉      | 19657/50000 [05:21<08:51, 57.13it/s]




 39%|███▉      | 19663/50000 [05:21<09:05, 55.60it/s]




 39%|███▉      | 19670/50000 [05:21<08:53, 56.85it/s]




 39%|███▉      | 19677/50000 [05:22<08:32, 59.22it/s]




 39%|███▉      | 19683/50000 [05:22<08:40, 58.28it/s]




 39%|███▉      | 19689/50000 [05:22<08:56, 56.51it/s]




 39%|███▉      | 19697/50000 [05:22<08:17, 60.88it/s]




 39%|███▉      | 19704/50000 [05:22<08:14, 61.24it/s]




 39%|███▉      | 19711/50000 [05:22<08:13, 61.34it/s]




 39%|███▉      | 19718/50000 [05:22<08:04, 62.56it/s]




 39%|███▉      | 19725/50000 [05:22<08:04, 62.42it/s]




 39%|███▉      | 19732/50000 [05:22<07:50, 64.39it/s]




 39%|███▉      | 19739/50000 [05:22<07:42, 65.47it/s]




 39%|███▉      | 19746/50000 [05:23<08:15, 61.06it/s]




 40%|███▉      | 19753/50000 [05:23<08:25, 59.80it/s]




 40%|███▉      | 19761/50000 [05:23<07:58, 63.22it/s]




 40%|███▉      | 19768/50000 [05:23<08:02, 62.68it/s]




 40%|███▉      | 19775/50000 [05:23<07:48, 64.48it/s]




 40%|███▉      | 19782/50000 [05:23<07:38, 65.91it/s]




 40%|███▉      | 19790/50000 [05:23<07:17, 69.10it/s]




 40%|███▉      | 19797/50000 [05:23<07:34, 66.46it/s]




 40%|███▉      | 19804/50000 [05:24<07:56, 63.33it/s]




 40%|███▉      | 19811/50000 [05:24<07:53, 63.82it/s]




 40%|███▉      | 19818/50000 [05:24<08:05, 62.12it/s]




 40%|███▉      | 19825/50000 [05:24<07:58, 63.12it/s]




 40%|███▉      | 19832/50000 [05:24<08:00, 62.82it/s]




 40%|███▉      | 19839/50000 [05:24<07:48, 64.32it/s]




 40%|███▉      | 19846/50000 [05:24<07:56, 63.29it/s]




 40%|███▉      | 19853/50000 [05:24<07:46, 64.67it/s]




 40%|███▉      | 19860/50000 [05:24<07:37, 65.87it/s]




 40%|███▉      | 19867/50000 [05:24<07:32, 66.54it/s]




 40%|███▉      | 19874/50000 [05:25<07:45, 64.69it/s]




 40%|███▉      | 19881/50000 [05:25<07:49, 64.16it/s]




 40%|███▉      | 19888/50000 [05:25<08:13, 61.04it/s]




 40%|███▉      | 19895/50000 [05:25<07:56, 63.18it/s]




 40%|███▉      | 19902/50000 [05:25<07:57, 63.03it/s]




 40%|███▉      | 19909/50000 [05:25<07:54, 63.43it/s]




 40%|███▉      | 19916/50000 [05:25<08:03, 62.19it/s]




 40%|███▉      | 19923/50000 [05:25<08:21, 59.94it/s]




 40%|███▉      | 19930/50000 [05:26<08:08, 61.53it/s]




 40%|███▉      | 19938/50000 [05:26<07:41, 65.18it/s]




 40%|███▉      | 19946/50000 [05:26<07:36, 65.83it/s]




 40%|███▉      | 19954/50000 [05:26<07:23, 67.72it/s]




 40%|███▉      | 19961/50000 [05:26<07:42, 64.95it/s]




 40%|███▉      | 19968/50000 [05:26<07:58, 62.78it/s]




 40%|███▉      | 19975/50000 [05:26<09:29, 52.77it/s]




 40%|███▉      | 19981/50000 [05:26<09:24, 53.19it/s]




 40%|███▉      | 19988/50000 [05:26<08:45, 57.08it/s]




 40%|███▉      | 19994/50000 [05:27<08:57, 55.87it/s]




 40%|████      | 20002/50000 [05:27<08:20, 59.95it/s]




 40%|████      | 20009/50000 [05:27<08:39, 57.74it/s]




 40%|████      | 20017/50000 [05:27<08:13, 60.75it/s]




 40%|████      | 20024/50000 [05:27<07:57, 62.79it/s]




 40%|████      | 20031/50000 [05:27<07:44, 64.57it/s]




 40%|████      | 20039/50000 [05:27<07:23, 67.63it/s]




 40%|████      | 20046/50000 [05:27<07:46, 64.26it/s]




 40%|████      | 20053/50000 [05:28<08:44, 57.09it/s]




 40%|████      | 20060/50000 [05:28<08:30, 58.70it/s]




 40%|████      | 20067/50000 [05:28<08:24, 59.30it/s]




 40%|████      | 20074/50000 [05:28<08:49, 56.49it/s]




 40%|████      | 20082/50000 [05:28<08:21, 59.64it/s]




 40%|████      | 20089/50000 [05:28<08:24, 59.29it/s]




 40%|████      | 20096/50000 [05:28<08:20, 59.80it/s]




 40%|████      | 20103/50000 [05:28<08:11, 60.79it/s]




 40%|████      | 20112/50000 [05:28<07:30, 66.35it/s]




 40%|████      | 20119/50000 [05:29<07:44, 64.30it/s]




 40%|████      | 20127/50000 [05:29<07:24, 67.25it/s]




 40%|████      | 20134/50000 [05:29<07:31, 66.18it/s]




 40%|████      | 20141/50000 [05:29<07:42, 64.55it/s]




 40%|████      | 20148/50000 [05:29<08:12, 60.65it/s]




 40%|████      | 20155/50000 [05:29<08:48, 56.51it/s]




 40%|████      | 20163/50000 [05:29<08:09, 60.99it/s]




 40%|████      | 20170/50000 [05:29<08:13, 60.45it/s]




 40%|████      | 20177/50000 [05:30<09:00, 55.19it/s]




 40%|████      | 20183/50000 [05:30<08:54, 55.80it/s]




 40%|████      | 20189/50000 [05:30<09:22, 52.97it/s]




 40%|████      | 20196/50000 [05:30<09:11, 54.01it/s]




 40%|████      | 20202/50000 [05:30<09:20, 53.21it/s]




 40%|████      | 20208/50000 [05:30<09:06, 54.52it/s]




 40%|████      | 20215/50000 [05:30<08:34, 57.85it/s]




 40%|████      | 20221/50000 [05:30<08:36, 57.69it/s]




 40%|████      | 20228/50000 [05:30<08:13, 60.31it/s]




 40%|████      | 20235/50000 [05:31<08:05, 61.32it/s]




 40%|████      | 20243/50000 [05:31<07:41, 64.55it/s]




 40%|████      | 20250/50000 [05:31<07:43, 64.15it/s]




 41%|████      | 20257/50000 [05:31<07:50, 63.18it/s]




 41%|████      | 20264/50000 [05:31<07:47, 63.62it/s]




 41%|████      | 20271/50000 [05:31<07:44, 64.02it/s]




 41%|████      | 20278/50000 [05:31<07:34, 65.39it/s]




 41%|████      | 20285/50000 [05:31<07:59, 61.98it/s]




 41%|████      | 20292/50000 [05:31<08:14, 60.11it/s]




 41%|████      | 20299/50000 [05:32<08:10, 60.54it/s]




 41%|████      | 20306/50000 [05:32<08:01, 61.65it/s]




 41%|████      | 20313/50000 [05:32<08:00, 61.79it/s]




 41%|████      | 20320/50000 [05:32<08:10, 60.52it/s]




 41%|████      | 20327/50000 [05:32<08:29, 58.26it/s]




 41%|████      | 20335/50000 [05:32<07:59, 61.91it/s]




 41%|████      | 20342/50000 [05:32<08:21, 59.15it/s]




 41%|████      | 20350/50000 [05:32<07:53, 62.58it/s]




 41%|████      | 20357/50000 [05:33<07:47, 63.37it/s]




 41%|████      | 20364/50000 [05:33<07:47, 63.33it/s]




 41%|████      | 20371/50000 [05:33<08:15, 59.74it/s]




 41%|████      | 20378/50000 [05:33<08:56, 55.16it/s]




 41%|████      | 20384/50000 [05:33<09:48, 50.33it/s]




 41%|████      | 20391/50000 [05:33<09:19, 52.88it/s]




 41%|████      | 20397/50000 [05:33<09:58, 49.46it/s]




 41%|████      | 20403/50000 [05:33<10:03, 49.05it/s]




 41%|████      | 20409/50000 [05:34<10:25, 47.28it/s]




 41%|████      | 20414/50000 [05:34<10:34, 46.62it/s]




 41%|████      | 20419/50000 [05:34<11:10, 44.10it/s]




 41%|████      | 20424/50000 [05:34<10:49, 45.51it/s]




 41%|████      | 20429/50000 [05:34<11:16, 43.74it/s]




 41%|████      | 20434/50000 [05:34<11:51, 41.57it/s]




 41%|████      | 20440/50000 [05:34<11:55, 41.29it/s]




 41%|████      | 20446/50000 [05:34<11:22, 43.32it/s]




 41%|████      | 20452/50000 [05:35<10:26, 47.18it/s]




 41%|████      | 20457/50000 [05:35<11:18, 43.52it/s]




 41%|████      | 20463/50000 [05:35<10:26, 47.13it/s]




 41%|████      | 20470/50000 [05:35<09:28, 51.93it/s]




 41%|████      | 20477/50000 [05:35<08:50, 55.66it/s]




 41%|████      | 20484/50000 [05:35<08:22, 58.71it/s]




 41%|████      | 20492/50000 [05:35<08:02, 61.21it/s]




 41%|████      | 20499/50000 [05:35<07:45, 63.38it/s]




 41%|████      | 20506/50000 [05:35<07:38, 64.32it/s]




 41%|████      | 20513/50000 [05:36<07:44, 63.44it/s]




 41%|████      | 20520/50000 [05:36<07:37, 64.42it/s]




 41%|████      | 20527/50000 [05:36<08:25, 58.31it/s]




 41%|████      | 20533/50000 [05:36<08:27, 58.02it/s]




 41%|████      | 20539/50000 [05:36<08:31, 57.56it/s]




 41%|████      | 20545/50000 [05:36<09:32, 51.43it/s]




 41%|████      | 20551/50000 [05:36<10:12, 48.10it/s]




 41%|████      | 20556/50000 [05:36<10:38, 46.14it/s]




 41%|████      | 20561/50000 [05:37<11:40, 42.04it/s]




 41%|████      | 20566/50000 [05:37<11:08, 44.06it/s]




 41%|████      | 20572/50000 [05:37<10:21, 47.34it/s]




 41%|████      | 20579/50000 [05:37<09:22, 52.35it/s]




 41%|████      | 20585/50000 [05:37<09:30, 51.53it/s]




 41%|████      | 20592/50000 [05:37<09:00, 54.43it/s]




 41%|████      | 20598/50000 [05:37<08:59, 54.51it/s]




 41%|████      | 20606/50000 [05:37<08:08, 60.17it/s]




 41%|████      | 20614/50000 [05:37<07:44, 63.28it/s]




 41%|████      | 20621/50000 [05:38<07:42, 63.57it/s]




 41%|████▏     | 20628/50000 [05:38<07:34, 64.65it/s]




 41%|████▏     | 20636/50000 [05:38<07:14, 67.60it/s]




 41%|████▏     | 20644/50000 [05:38<07:00, 69.84it/s]




 41%|████▏     | 20652/50000 [05:38<07:10, 68.22it/s]




 41%|████▏     | 20659/50000 [05:38<07:18, 66.93it/s]




 41%|████▏     | 20667/50000 [05:38<07:14, 67.58it/s]




 41%|████▏     | 20674/50000 [05:38<07:25, 65.84it/s]




 41%|████▏     | 20683/50000 [05:38<06:58, 69.98it/s]




 41%|████▏     | 20691/50000 [05:39<07:03, 69.19it/s]




 41%|████▏     | 20698/50000 [05:39<07:33, 64.68it/s]




 41%|████▏     | 20705/50000 [05:39<07:29, 65.14it/s]




 41%|████▏     | 20714/50000 [05:39<07:00, 69.59it/s]




 41%|████▏     | 20722/50000 [05:39<07:25, 65.70it/s]




 41%|████▏     | 20729/50000 [05:39<07:39, 63.70it/s]




 41%|████▏     | 20737/50000 [05:39<07:20, 66.38it/s]




 41%|████▏     | 20744/50000 [05:39<07:18, 66.71it/s]




 42%|████▏     | 20752/50000 [05:39<06:59, 69.71it/s]




 42%|████▏     | 20760/50000 [05:40<07:19, 66.60it/s]




 42%|████▏     | 20767/50000 [05:40<07:34, 64.38it/s]




 42%|████▏     | 20775/50000 [05:40<07:16, 66.89it/s]




 42%|████▏     | 20782/50000 [05:40<07:20, 66.31it/s]




 42%|████▏     | 20790/50000 [05:40<07:08, 68.16it/s]




 42%|████▏     | 20797/50000 [05:40<07:17, 66.79it/s]




 42%|████▏     | 20804/50000 [05:40<07:14, 67.19it/s]




 42%|████▏     | 20811/50000 [05:40<07:13, 67.28it/s]




 42%|████▏     | 20819/50000 [05:40<07:02, 69.06it/s]




 42%|████▏     | 20826/50000 [05:41<07:07, 68.18it/s]




 42%|████▏     | 20834/50000 [05:41<06:52, 70.64it/s]




 42%|████▏     | 20842/50000 [05:41<07:10, 67.70it/s]




 42%|████▏     | 20849/50000 [05:41<07:34, 64.19it/s]




 42%|████▏     | 20857/50000 [05:41<07:22, 65.93it/s]




 42%|████▏     | 20864/50000 [05:41<07:25, 65.46it/s]




 42%|████▏     | 20872/50000 [05:41<07:02, 68.92it/s]




 42%|████▏     | 20880/50000 [05:41<06:56, 69.89it/s]




 42%|████▏     | 20888/50000 [05:41<06:43, 72.11it/s]




 42%|████▏     | 20896/50000 [05:42<06:32, 74.16it/s]




 42%|████▏     | 20904/50000 [05:42<06:38, 72.98it/s]




 42%|████▏     | 20912/50000 [05:42<06:31, 74.38it/s]




 42%|████▏     | 20920/50000 [05:42<06:50, 70.81it/s]




 42%|████▏     | 20928/50000 [05:42<07:03, 68.68it/s]




 42%|████▏     | 20936/50000 [05:42<06:48, 71.20it/s]




 42%|████▏     | 20944/50000 [05:42<07:00, 69.11it/s]




 42%|████▏     | 20951/50000 [05:42<07:04, 68.35it/s]




 42%|████▏     | 20958/50000 [05:42<07:23, 65.48it/s]




 42%|████▏     | 20965/50000 [05:43<07:23, 65.51it/s]




 42%|████▏     | 20972/50000 [05:43<07:19, 66.09it/s]




 42%|████▏     | 20979/50000 [05:43<07:30, 64.40it/s]




 42%|████▏     | 20986/50000 [05:43<07:23, 65.41it/s]




 42%|████▏     | 20993/50000 [05:43<07:25, 65.10it/s]




 42%|████▏     | 21000/50000 [05:43<07:26, 64.89it/s]




 42%|████▏     | 21007/50000 [05:43<07:32, 64.03it/s]




 42%|████▏     | 21014/50000 [05:43<08:22, 57.66it/s]




 42%|████▏     | 21020/50000 [05:43<08:45, 55.17it/s]




 42%|████▏     | 21027/50000 [05:44<08:26, 57.23it/s]




 42%|████▏     | 21034/50000 [05:44<08:08, 59.35it/s]




 42%|████▏     | 21041/50000 [05:44<07:58, 60.47it/s]




 42%|████▏     | 21049/50000 [05:44<07:26, 64.81it/s]




 42%|████▏     | 21057/50000 [05:44<07:08, 67.54it/s]




 42%|████▏     | 21064/50000 [05:44<07:27, 64.72it/s]




 42%|████▏     | 21071/50000 [05:44<07:51, 61.40it/s]




 42%|████▏     | 21079/50000 [05:44<07:29, 64.36it/s]




 42%|████▏     | 21086/50000 [05:44<07:22, 65.36it/s]




 42%|████▏     | 21093/50000 [05:45<07:33, 63.73it/s]




 42%|████▏     | 21100/50000 [05:45<07:27, 64.54it/s]




 42%|████▏     | 21108/50000 [05:45<07:13, 66.68it/s]




 42%|████▏     | 21115/50000 [05:45<07:12, 66.73it/s]




 42%|████▏     | 21122/50000 [05:45<07:32, 63.85it/s]




 42%|████▏     | 21129/50000 [05:45<07:28, 64.36it/s]




 42%|████▏     | 21136/50000 [05:45<07:23, 65.09it/s]




 42%|████▏     | 21145/50000 [05:45<06:56, 69.23it/s]




 42%|████▏     | 21153/50000 [05:45<07:17, 65.96it/s]




 42%|████▏     | 21160/50000 [05:46<07:37, 63.01it/s]




 42%|████▏     | 21167/50000 [05:46<07:35, 63.33it/s]




 42%|████▏     | 21174/50000 [05:46<07:38, 62.82it/s]




 42%|████▏     | 21181/50000 [05:46<07:48, 61.45it/s]




 42%|████▏     | 21188/50000 [05:46<07:51, 61.17it/s]




 42%|████▏     | 21196/50000 [05:46<07:27, 64.43it/s]




 42%|████▏     | 21203/50000 [05:46<07:36, 63.03it/s]




 42%|████▏     | 21210/50000 [05:46<07:43, 62.09it/s]




 42%|████▏     | 21217/50000 [05:47<07:59, 60.02it/s]




 42%|████▏     | 21224/50000 [05:47<07:49, 61.27it/s]




 42%|████▏     | 21232/50000 [05:47<07:34, 63.29it/s]




 42%|████▏     | 21239/50000 [05:47<07:25, 64.49it/s]




 42%|████▏     | 21246/50000 [05:47<07:46, 61.58it/s]




 43%|████▎     | 21253/50000 [05:47<07:43, 62.06it/s]




 43%|████▎     | 21260/50000 [05:47<07:29, 63.95it/s]




 43%|████▎     | 21268/50000 [05:47<07:09, 66.90it/s]




 43%|████▎     | 21275/50000 [05:47<07:13, 66.30it/s]




 43%|████▎     | 21283/50000 [05:48<06:56, 68.95it/s]




 43%|████▎     | 21291/50000 [05:48<06:45, 70.74it/s]




 43%|████▎     | 21299/50000 [05:48<07:04, 67.67it/s]




 43%|████▎     | 21306/50000 [05:48<07:08, 66.94it/s]




 43%|████▎     | 21314/50000 [05:48<06:56, 68.80it/s]




 43%|████▎     | 21321/50000 [05:48<07:13, 66.09it/s]




 43%|████▎     | 21328/50000 [05:48<07:09, 66.69it/s]




 43%|████▎     | 21337/50000 [05:48<06:43, 70.99it/s]




 43%|████▎     | 21345/50000 [05:48<07:16, 65.59it/s]




 43%|████▎     | 21353/50000 [05:49<07:06, 67.12it/s]




 43%|████▎     | 21360/50000 [05:49<07:14, 65.90it/s]




 43%|████▎     | 21367/50000 [05:49<07:37, 62.64it/s]




 43%|████▎     | 21374/50000 [05:49<07:40, 62.21it/s]




 43%|████▎     | 21381/50000 [05:49<08:32, 55.80it/s]




 43%|████▎     | 21387/50000 [05:49<09:06, 52.32it/s]




 43%|████▎     | 21393/50000 [05:49<09:10, 51.98it/s]




 43%|████▎     | 21399/50000 [05:49<09:00, 52.90it/s]




 43%|████▎     | 21405/50000 [05:50<09:39, 49.35it/s]




 43%|████▎     | 21411/50000 [05:50<10:05, 47.25it/s]




 43%|████▎     | 21417/50000 [05:50<09:31, 50.00it/s]




 43%|████▎     | 21423/50000 [05:50<09:38, 49.42it/s]




 43%|████▎     | 21429/50000 [05:50<09:31, 50.00it/s]




 43%|████▎     | 21435/50000 [05:50<09:36, 49.55it/s]




 43%|████▎     | 21441/50000 [05:50<09:09, 51.97it/s]




 43%|████▎     | 21447/50000 [05:50<09:08, 52.08it/s]




 43%|████▎     | 21453/50000 [05:50<08:54, 53.40it/s]




 43%|████▎     | 21459/50000 [05:51<08:47, 54.07it/s]




 43%|████▎     | 21465/50000 [05:51<08:43, 54.47it/s]




 43%|████▎     | 21471/50000 [05:51<08:34, 55.44it/s]




 43%|████▎     | 21478/50000 [05:51<08:06, 58.65it/s]




 43%|████▎     | 21485/50000 [05:51<07:56, 59.80it/s]




 43%|████▎     | 21492/50000 [05:51<07:52, 60.32it/s]




 43%|████▎     | 21499/50000 [05:51<07:35, 62.64it/s]




 43%|████▎     | 21506/50000 [05:51<07:29, 63.33it/s]




 43%|████▎     | 21513/50000 [05:51<07:23, 64.17it/s]




 43%|████▎     | 21520/50000 [05:52<07:29, 63.36it/s]




 43%|████▎     | 21527/50000 [05:52<07:43, 61.49it/s]




 43%|████▎     | 21534/50000 [05:52<07:51, 60.40it/s]




 43%|████▎     | 21541/50000 [05:52<08:20, 56.83it/s]




 43%|████▎     | 21547/50000 [05:52<08:49, 53.76it/s]




 43%|████▎     | 21553/50000 [05:52<08:59, 52.76it/s]




 43%|████▎     | 21559/50000 [05:52<09:11, 51.54it/s]




 43%|████▎     | 21565/50000 [05:52<09:45, 48.60it/s]




 43%|████▎     | 21571/50000 [05:53<09:17, 50.98it/s]




 43%|████▎     | 21578/50000 [05:53<08:35, 55.15it/s]




 43%|████▎     | 21586/50000 [05:53<07:57, 59.50it/s]




 43%|████▎     | 21593/50000 [05:53<07:52, 60.10it/s]




 43%|████▎     | 21600/50000 [05:53<07:34, 62.48it/s]




 43%|████▎     | 21607/50000 [05:53<07:25, 63.73it/s]




 43%|████▎     | 21614/50000 [05:53<07:31, 62.89it/s]




 43%|████▎     | 21621/50000 [05:53<07:20, 64.38it/s]




 43%|████▎     | 21628/50000 [05:53<07:12, 65.65it/s]




 43%|████▎     | 21636/50000 [05:54<06:52, 68.72it/s]




 43%|████▎     | 21643/50000 [05:54<07:16, 64.93it/s]




 43%|████▎     | 21651/50000 [05:54<06:55, 68.16it/s]




 43%|████▎     | 21659/50000 [05:54<06:46, 69.79it/s]




 43%|████▎     | 21667/50000 [05:54<06:56, 68.09it/s]




 43%|████▎     | 21674/50000 [05:54<07:01, 67.13it/s]




 43%|████▎     | 21681/50000 [05:54<07:08, 66.10it/s]




 43%|████▎     | 21688/50000 [05:54<07:08, 66.13it/s]




 43%|████▎     | 21695/50000 [05:54<07:18, 64.51it/s]




 43%|████▎     | 21702/50000 [05:55<07:29, 62.92it/s]




 43%|████▎     | 21709/50000 [05:55<07:30, 62.84it/s]




 43%|████▎     | 21716/50000 [05:55<07:46, 60.67it/s]




 43%|████▎     | 21723/50000 [05:55<07:41, 61.26it/s]




 43%|████▎     | 21731/50000 [05:55<07:16, 64.81it/s]




 43%|████▎     | 21738/50000 [05:55<07:31, 62.61it/s]




 43%|████▎     | 21745/50000 [05:55<07:33, 62.29it/s]




 44%|████▎     | 21753/50000 [05:55<07:08, 65.94it/s]




 44%|████▎     | 21760/50000 [05:55<07:18, 64.41it/s]




 44%|████▎     | 21767/50000 [05:56<07:46, 60.55it/s]




 44%|████▎     | 21774/50000 [05:56<07:36, 61.89it/s]




 44%|████▎     | 21782/50000 [05:56<07:12, 65.24it/s]




 44%|████▎     | 21789/50000 [05:56<07:28, 62.97it/s]




 44%|████▎     | 21796/50000 [05:56<07:14, 64.88it/s]




 44%|████▎     | 21803/50000 [05:56<07:30, 62.65it/s]




 44%|████▎     | 21811/50000 [05:56<07:15, 64.78it/s]




 44%|████▎     | 21818/50000 [05:56<07:33, 62.09it/s]




 44%|████▎     | 21825/50000 [05:56<07:22, 63.62it/s]




 44%|████▎     | 21832/50000 [05:57<07:44, 60.70it/s]




 44%|████▎     | 21839/50000 [05:57<07:28, 62.76it/s]




 44%|████▎     | 21846/50000 [05:57<07:25, 63.24it/s]




 44%|████▎     | 21853/50000 [05:57<07:28, 62.73it/s]




 44%|████▎     | 21861/50000 [05:57<07:08, 65.63it/s]




 44%|████▎     | 21868/50000 [05:57<07:03, 66.37it/s]




 44%|████▍     | 21875/50000 [05:57<07:02, 66.62it/s]




 44%|████▍     | 21882/50000 [05:57<07:26, 63.03it/s]




 44%|████▍     | 21889/50000 [05:57<07:14, 64.75it/s]




 44%|████▍     | 21896/50000 [05:58<07:34, 61.79it/s]




 44%|████▍     | 21903/50000 [05:58<07:23, 63.34it/s]




 44%|████▍     | 21910/50000 [05:58<07:20, 63.83it/s]




 44%|████▍     | 21917/50000 [05:58<07:18, 63.99it/s]




 44%|████▍     | 21924/50000 [05:58<07:21, 63.59it/s]




 44%|████▍     | 21932/50000 [05:58<07:01, 66.62it/s]




 44%|████▍     | 21939/50000 [05:58<07:22, 63.44it/s]




 44%|████▍     | 21947/50000 [05:58<06:57, 67.17it/s]




 44%|████▍     | 21954/50000 [05:58<07:28, 62.59it/s]




 44%|████▍     | 21961/50000 [05:59<07:25, 62.95it/s]




 44%|████▍     | 21969/50000 [05:59<07:04, 65.97it/s]




 44%|████▍     | 21976/50000 [05:59<06:58, 66.99it/s]




 44%|████▍     | 21984/50000 [05:59<06:42, 69.56it/s]




 44%|████▍     | 21992/50000 [05:59<06:53, 67.68it/s]




 44%|████▍     | 21999/50000 [05:59<07:09, 65.17it/s]




 44%|████▍     | 22007/50000 [05:59<06:54, 67.49it/s]




 44%|████▍     | 22014/50000 [05:59<07:17, 63.98it/s]




 44%|████▍     | 22021/50000 [05:59<07:06, 65.54it/s]




 44%|████▍     | 22029/50000 [06:00<06:50, 68.11it/s]




 44%|████▍     | 22038/50000 [06:00<06:25, 72.46it/s]




 44%|████▍     | 22047/50000 [06:00<06:15, 74.34it/s]




 44%|████▍     | 22055/50000 [06:00<06:28, 71.92it/s]




 44%|████▍     | 22063/50000 [06:00<06:38, 70.14it/s]




 44%|████▍     | 22071/50000 [06:00<06:43, 69.29it/s]




 44%|████▍     | 22078/50000 [06:00<06:55, 67.16it/s]




 44%|████▍     | 22085/50000 [06:00<07:03, 65.96it/s]




 44%|████▍     | 22092/50000 [06:00<07:05, 65.61it/s]




 44%|████▍     | 22099/50000 [06:01<07:20, 63.30it/s]




 44%|████▍     | 22107/50000 [06:01<07:06, 65.42it/s]




 44%|████▍     | 22116/50000 [06:01<06:41, 69.49it/s]




 44%|████▍     | 22124/50000 [06:01<06:54, 67.29it/s]




 44%|████▍     | 22131/50000 [06:01<07:05, 65.47it/s]




 44%|████▍     | 22138/50000 [06:01<07:38, 60.74it/s]




 44%|████▍     | 22146/50000 [06:01<07:34, 61.31it/s]




 44%|████▍     | 22153/50000 [06:01<07:32, 61.55it/s]




 44%|████▍     | 22160/50000 [06:02<07:26, 62.37it/s]




 44%|████▍     | 22167/50000 [06:02<07:52, 58.85it/s]




 44%|████▍     | 22173/50000 [06:02<07:58, 58.20it/s]




 44%|████▍     | 22180/50000 [06:02<07:36, 60.95it/s]




 44%|████▍     | 22187/50000 [06:02<07:40, 60.42it/s]




 44%|████▍     | 22194/50000 [06:02<07:28, 62.00it/s]




 44%|████▍     | 22201/50000 [06:02<07:32, 61.46it/s]




 44%|████▍     | 22208/50000 [06:02<07:33, 61.30it/s]




 44%|████▍     | 22215/50000 [06:02<07:43, 59.97it/s]




 44%|████▍     | 22222/50000 [06:03<07:35, 60.92it/s]




 44%|████▍     | 22231/50000 [06:03<06:56, 66.60it/s]




 44%|████▍     | 22238/50000 [06:03<06:52, 67.25it/s]




 44%|████▍     | 22245/50000 [06:03<07:12, 64.18it/s]




 45%|████▍     | 22252/50000 [06:03<07:20, 63.03it/s]




 45%|████▍     | 22260/50000 [06:03<07:02, 65.71it/s]




 45%|████▍     | 22267/50000 [06:03<07:26, 62.18it/s]




 45%|████▍     | 22274/50000 [06:03<07:22, 62.66it/s]




 45%|████▍     | 22281/50000 [06:03<07:29, 61.68it/s]




 45%|████▍     | 22288/50000 [06:04<07:29, 61.65it/s]




 45%|████▍     | 22295/50000 [06:04<07:29, 61.62it/s]




 45%|████▍     | 22302/50000 [06:04<07:18, 63.17it/s]




 45%|████▍     | 22309/50000 [06:04<07:24, 62.32it/s]




 45%|████▍     | 22316/50000 [06:04<07:21, 62.76it/s]




 45%|████▍     | 22323/50000 [06:04<07:09, 64.37it/s]




 45%|████▍     | 22330/50000 [06:04<07:11, 64.20it/s]




 45%|████▍     | 22337/50000 [06:04<07:14, 63.73it/s]




 45%|████▍     | 22344/50000 [06:04<07:05, 64.99it/s]




 45%|████▍     | 22352/50000 [06:05<06:45, 68.21it/s]




 45%|████▍     | 22359/50000 [06:05<06:54, 66.64it/s]




 45%|████▍     | 22366/50000 [06:05<06:53, 66.89it/s]




 45%|████▍     | 22373/50000 [06:05<07:15, 63.43it/s]




 45%|████▍     | 22380/50000 [06:05<08:06, 56.83it/s]




 45%|████▍     | 22386/50000 [06:05<08:55, 51.55it/s]




 45%|████▍     | 22392/50000 [06:05<09:13, 49.85it/s]




 45%|████▍     | 22398/50000 [06:05<09:19, 49.32it/s]




 45%|████▍     | 22404/50000 [06:06<09:37, 47.79it/s]




 45%|████▍     | 22410/50000 [06:06<09:31, 48.24it/s]




 45%|████▍     | 22415/50000 [06:06<09:53, 46.48it/s]




 45%|████▍     | 22421/50000 [06:06<09:16, 49.52it/s]




 45%|████▍     | 22427/50000 [06:06<08:55, 51.49it/s]




 45%|████▍     | 22433/50000 [06:06<08:51, 51.87it/s]




 45%|████▍     | 22439/50000 [06:06<08:48, 52.13it/s]




 45%|████▍     | 22445/50000 [06:06<08:56, 51.39it/s]




 45%|████▍     | 22451/50000 [06:07<08:43, 52.61it/s]




 45%|████▍     | 22457/50000 [06:07<08:36, 53.36it/s]




 45%|████▍     | 22464/50000 [06:07<08:05, 56.72it/s]




 45%|████▍     | 22471/50000 [06:07<07:56, 57.80it/s]




 45%|████▍     | 22478/50000 [06:07<07:36, 60.25it/s]




 45%|████▍     | 22486/50000 [06:07<07:19, 62.67it/s]




 45%|████▍     | 22493/50000 [06:07<07:43, 59.40it/s]




 45%|████▌     | 22500/50000 [06:07<08:00, 57.21it/s]




 45%|████▌     | 22507/50000 [06:07<07:36, 60.26it/s]




 45%|████▌     | 22514/50000 [06:08<07:23, 61.93it/s]




 45%|████▌     | 22521/50000 [06:08<08:11, 55.86it/s]




 45%|████▌     | 22527/50000 [06:08<08:28, 54.03it/s]




 45%|████▌     | 22533/50000 [06:08<08:46, 52.17it/s]




 45%|████▌     | 22539/50000 [06:08<08:52, 51.54it/s]




 45%|████▌     | 22545/50000 [06:08<09:17, 49.23it/s]




 45%|████▌     | 22551/50000 [06:08<09:22, 48.77it/s]




 45%|████▌     | 22556/50000 [06:08<09:34, 47.77it/s]




 45%|████▌     | 22563/50000 [06:09<09:00, 50.77it/s]




 45%|████▌     | 22569/50000 [06:09<08:40, 52.70it/s]




 45%|████▌     | 22576/50000 [06:09<08:14, 55.47it/s]




 45%|████▌     | 22583/50000 [06:09<07:51, 58.19it/s]




 45%|████▌     | 22589/50000 [06:09<07:54, 57.75it/s]




 45%|████▌     | 22597/50000 [06:09<07:26, 61.41it/s]




 45%|████▌     | 22606/50000 [06:09<06:49, 66.97it/s]




 45%|████▌     | 22613/50000 [06:09<07:07, 64.09it/s]




 45%|████▌     | 22621/50000 [06:09<06:45, 67.46it/s]




 45%|████▌     | 22628/50000 [06:10<06:56, 65.76it/s]




 45%|████▌     | 22636/50000 [06:10<06:42, 67.93it/s]




 45%|████▌     | 22643/50000 [06:10<06:52, 66.26it/s]




 45%|████▌     | 22650/50000 [06:10<06:57, 65.51it/s]




 45%|████▌     | 22657/50000 [06:10<07:01, 64.81it/s]




 45%|████▌     | 22664/50000 [06:10<07:14, 62.94it/s]




 45%|████▌     | 22671/50000 [06:10<07:11, 63.37it/s]




 45%|████▌     | 22678/50000 [06:10<07:08, 63.75it/s]




 45%|████▌     | 22685/50000 [06:10<07:01, 64.83it/s]




 45%|████▌     | 22693/50000 [06:11<06:48, 66.89it/s]




 45%|████▌     | 22700/50000 [06:11<06:52, 66.12it/s]




 45%|████▌     | 22707/50000 [06:11<07:01, 64.69it/s]




 45%|████▌     | 22714/50000 [06:11<07:36, 59.81it/s]




 45%|████▌     | 22721/50000 [06:11<09:00, 50.44it/s]




 45%|████▌     | 22727/50000 [06:11<09:11, 49.48it/s]




 45%|████▌     | 22735/50000 [06:11<08:15, 54.99it/s]




 45%|████▌     | 22742/50000 [06:11<07:52, 57.65it/s]




 46%|████▌     | 22750/50000 [06:12<07:22, 61.51it/s]




 46%|████▌     | 22757/50000 [06:12<07:28, 60.73it/s]




 46%|████▌     | 22764/50000 [06:12<07:23, 61.46it/s]




 46%|████▌     | 22771/50000 [06:12<07:08, 63.49it/s]




 46%|████▌     | 22778/50000 [06:12<07:42, 58.85it/s]




 46%|████▌     | 22785/50000 [06:12<07:27, 60.88it/s]




 46%|████▌     | 22792/50000 [06:12<07:12, 62.96it/s]




 46%|████▌     | 22799/50000 [06:12<07:13, 62.71it/s]




 46%|████▌     | 22807/50000 [06:12<07:00, 64.67it/s]




 46%|████▌     | 22815/50000 [06:13<06:41, 67.79it/s]




 46%|████▌     | 22822/50000 [06:13<06:51, 66.07it/s]




 46%|████▌     | 22829/50000 [06:13<07:04, 63.94it/s]




 46%|████▌     | 22836/50000 [06:13<07:23, 61.23it/s]




 46%|████▌     | 22843/50000 [06:13<07:14, 62.48it/s]




 46%|████▌     | 22850/50000 [06:13<07:11, 62.87it/s]




 46%|████▌     | 22857/50000 [06:13<07:02, 64.19it/s]




 46%|████▌     | 22865/50000 [06:13<06:41, 67.59it/s]




 46%|████▌     | 22872/50000 [06:13<06:51, 65.85it/s]




 46%|████▌     | 22879/50000 [06:14<07:01, 64.32it/s]




 46%|████▌     | 22886/50000 [06:14<07:06, 63.64it/s]




 46%|████▌     | 22893/50000 [06:14<07:22, 61.19it/s]




 46%|████▌     | 22902/50000 [06:14<06:54, 65.44it/s]




 46%|████▌     | 22909/50000 [06:14<07:18, 61.85it/s]




 46%|████▌     | 22916/50000 [06:14<07:12, 62.67it/s]




 46%|████▌     | 22923/50000 [06:14<07:03, 63.87it/s]




 46%|████▌     | 22930/50000 [06:14<06:59, 64.55it/s]




 46%|████▌     | 22937/50000 [06:14<06:59, 64.51it/s]




 46%|████▌     | 22944/50000 [06:15<07:19, 61.59it/s]




 46%|████▌     | 22952/50000 [06:15<06:56, 64.91it/s]




 46%|████▌     | 22959/50000 [06:15<06:54, 65.30it/s]




 46%|████▌     | 22966/50000 [06:15<07:15, 62.09it/s]




 46%|████▌     | 22973/50000 [06:15<07:23, 60.96it/s]




 46%|████▌     | 22980/50000 [06:15<07:06, 63.29it/s]




 46%|████▌     | 22987/50000 [06:15<07:16, 61.92it/s]




 46%|████▌     | 22994/50000 [06:15<07:22, 61.08it/s]




 46%|████▌     | 23001/50000 [06:15<07:26, 60.43it/s]




 46%|████▌     | 23008/50000 [06:16<07:12, 62.39it/s]




 46%|████▌     | 23015/50000 [06:16<07:23, 60.85it/s]




 46%|████▌     | 23023/50000 [06:16<06:58, 64.49it/s]




 46%|████▌     | 23030/50000 [06:16<06:49, 65.91it/s]




 46%|████▌     | 23037/50000 [06:16<06:49, 65.82it/s]




 46%|████▌     | 23044/50000 [06:16<06:47, 66.12it/s]




 46%|████▌     | 23052/50000 [06:16<06:35, 68.20it/s]




 46%|████▌     | 23059/50000 [06:16<06:36, 67.98it/s]




 46%|████▌     | 23067/50000 [06:16<06:25, 69.94it/s]




 46%|████▌     | 23075/50000 [06:17<06:20, 70.81it/s]




 46%|████▌     | 23083/50000 [06:17<06:24, 70.02it/s]




 46%|████▌     | 23091/50000 [06:17<06:34, 68.21it/s]




 46%|████▌     | 23098/50000 [06:17<06:46, 66.26it/s]




 46%|████▌     | 23105/50000 [06:17<06:58, 64.25it/s]




 46%|████▌     | 23112/50000 [06:17<07:07, 62.91it/s]




 46%|████▌     | 23119/50000 [06:17<07:00, 63.86it/s]




 46%|████▋     | 23127/50000 [06:17<06:36, 67.85it/s]




 46%|████▋     | 23134/50000 [06:17<06:34, 68.14it/s]




 46%|████▋     | 23141/50000 [06:18<07:05, 63.17it/s]




 46%|████▋     | 23148/50000 [06:18<06:54, 64.76it/s]




 46%|████▋     | 23155/50000 [06:18<07:03, 63.42it/s]




 46%|████▋     | 23163/50000 [06:18<06:43, 66.50it/s]




 46%|████▋     | 23170/50000 [06:18<06:57, 64.22it/s]




 46%|████▋     | 23177/50000 [06:18<06:54, 64.72it/s]




 46%|████▋     | 23184/50000 [06:18<06:51, 65.16it/s]




 46%|████▋     | 23191/50000 [06:18<07:05, 63.00it/s]




 46%|████▋     | 23198/50000 [06:18<07:02, 63.41it/s]




 46%|████▋     | 23205/50000 [06:19<07:20, 60.88it/s]




 46%|████▋     | 23212/50000 [06:19<07:10, 62.22it/s]




 46%|████▋     | 23220/50000 [06:19<06:50, 65.25it/s]




 46%|████▋     | 23227/50000 [06:19<06:43, 66.28it/s]




 46%|████▋     | 23235/50000 [06:19<06:29, 68.66it/s]




 46%|████▋     | 23242/50000 [06:19<06:48, 65.44it/s]




 46%|████▋     | 23249/50000 [06:19<06:49, 65.30it/s]




 47%|████▋     | 23256/50000 [06:19<06:48, 65.39it/s]




 47%|████▋     | 23263/50000 [06:19<06:42, 66.38it/s]




 47%|████▋     | 23270/50000 [06:20<06:48, 65.40it/s]




 47%|████▋     | 23277/50000 [06:20<06:56, 64.11it/s]




 47%|████▋     | 23285/50000 [06:20<06:45, 65.87it/s]




 47%|████▋     | 23292/50000 [06:20<07:07, 62.45it/s]




 47%|████▋     | 23299/50000 [06:20<07:36, 58.45it/s]




 47%|████▋     | 23306/50000 [06:20<07:19, 60.74it/s]




 47%|████▋     | 23313/50000 [06:20<07:07, 62.45it/s]




 47%|████▋     | 23321/50000 [06:20<06:50, 64.95it/s]




 47%|████▋     | 23328/50000 [06:20<07:14, 61.38it/s]




 47%|████▋     | 23335/50000 [06:21<07:11, 61.76it/s]




 47%|████▋     | 23342/50000 [06:21<07:13, 61.54it/s]




 47%|████▋     | 23350/50000 [06:21<06:59, 63.49it/s]




 47%|████▋     | 23357/50000 [06:21<07:11, 61.74it/s]




 47%|████▋     | 23364/50000 [06:21<07:47, 57.03it/s]




 47%|████▋     | 23370/50000 [06:21<08:13, 53.99it/s]




 47%|████▋     | 23376/50000 [06:21<08:44, 50.77it/s]




 47%|████▋     | 23382/50000 [06:21<09:11, 48.26it/s]




 47%|████▋     | 23387/50000 [06:22<09:51, 45.00it/s]




 47%|████▋     | 23392/50000 [06:22<09:43, 45.60it/s]




 47%|████▋     | 23397/50000 [06:22<09:43, 45.59it/s]




 47%|████▋     | 23402/50000 [06:22<09:57, 44.50it/s]




 47%|████▋     | 23408/50000 [06:22<09:45, 45.42it/s]




 47%|████▋     | 23414/50000 [06:22<09:19, 47.52it/s]




 47%|████▋     | 23419/50000 [06:22<09:31, 46.52it/s]




 47%|████▋     | 23424/50000 [06:22<09:37, 45.98it/s]




 47%|████▋     | 23430/50000 [06:23<09:03, 48.88it/s]




 47%|████▋     | 23436/50000 [06:23<08:34, 51.59it/s]




 47%|████▋     | 23444/50000 [06:23<07:46, 56.89it/s]




 47%|████▋     | 23452/50000 [06:23<07:15, 61.02it/s]




 47%|████▋     | 23459/50000 [06:23<07:17, 60.62it/s]




 47%|████▋     | 23466/50000 [06:23<07:15, 60.90it/s]




 47%|████▋     | 23473/50000 [06:23<07:08, 61.91it/s]




 47%|████▋     | 23480/50000 [06:23<06:54, 64.00it/s]




 47%|████▋     | 23487/50000 [06:23<06:50, 64.65it/s]




 47%|████▋     | 23494/50000 [06:24<06:58, 63.26it/s]




 47%|████▋     | 23501/50000 [06:24<07:49, 56.39it/s]




 47%|████▋     | 23508/50000 [06:24<07:26, 59.36it/s]




 47%|████▋     | 23515/50000 [06:24<07:31, 58.72it/s]




 47%|████▋     | 23521/50000 [06:24<08:11, 53.90it/s]




 47%|████▋     | 23527/50000 [06:24<08:24, 52.44it/s]




 47%|████▋     | 23533/50000 [06:24<09:31, 46.34it/s]




 47%|████▋     | 23538/50000 [06:24<09:24, 46.89it/s]




 47%|████▋     | 23544/50000 [06:25<08:50, 49.84it/s]




 47%|████▋     | 23550/50000 [06:25<08:25, 52.33it/s]




 47%|████▋     | 23557/50000 [06:25<07:58, 55.23it/s]




 47%|████▋     | 23564/50000 [06:25<07:38, 57.70it/s]




 47%|████▋     | 23571/50000 [06:25<07:15, 60.74it/s]




 47%|████▋     | 23578/50000 [06:25<07:12, 61.06it/s]




 47%|████▋     | 23585/50000 [06:25<07:00, 62.86it/s]




 47%|████▋     | 23592/50000 [06:25<07:18, 60.21it/s]




 47%|████▋     | 23599/50000 [06:25<07:20, 59.99it/s]




 47%|████▋     | 23606/50000 [06:25<07:07, 61.73it/s]




 47%|████▋     | 23613/50000 [06:26<06:59, 62.84it/s]




 47%|████▋     | 23620/50000 [06:26<06:56, 63.30it/s]




 47%|████▋     | 23627/50000 [06:26<06:48, 64.50it/s]




 47%|████▋     | 23634/50000 [06:26<06:51, 64.11it/s]




 47%|████▋     | 23641/50000 [06:26<06:55, 63.50it/s]




 47%|████▋     | 23648/50000 [06:26<06:50, 64.20it/s]




 47%|████▋     | 23655/50000 [06:26<06:49, 64.33it/s]




 47%|████▋     | 23662/50000 [06:26<06:44, 65.07it/s]




 47%|████▋     | 23669/50000 [06:26<06:54, 63.46it/s]




 47%|████▋     | 23676/50000 [06:27<06:58, 62.88it/s]




 47%|████▋     | 23683/50000 [06:27<06:51, 64.02it/s]




 47%|████▋     | 23690/50000 [06:27<06:45, 64.84it/s]




 47%|████▋     | 23697/50000 [06:27<06:39, 65.80it/s]




 47%|████▋     | 23704/50000 [06:27<06:54, 63.42it/s]




 47%|████▋     | 23711/50000 [06:27<07:02, 62.18it/s]




 47%|████▋     | 23718/50000 [06:27<07:01, 62.33it/s]




 47%|████▋     | 23725/50000 [06:27<06:53, 63.62it/s]




 47%|████▋     | 23733/50000 [06:27<06:30, 67.32it/s]




 47%|████▋     | 23741/50000 [06:28<06:16, 69.71it/s]




 47%|████▋     | 23749/50000 [06:28<06:05, 71.78it/s]




 48%|████▊     | 23757/50000 [06:28<06:17, 69.44it/s]




 48%|████▊     | 23765/50000 [06:28<06:25, 68.11it/s]




 48%|████▊     | 23772/50000 [06:28<06:38, 65.82it/s]




 48%|████▊     | 23780/50000 [06:28<06:18, 69.32it/s]




 48%|████▊     | 23788/50000 [06:28<06:16, 69.53it/s]




 48%|████▊     | 23796/50000 [06:28<06:33, 66.65it/s]




 48%|████▊     | 23804/50000 [06:28<06:15, 69.84it/s]




 48%|████▊     | 23812/50000 [06:29<06:31, 66.85it/s]




 48%|████▊     | 23819/50000 [06:29<06:31, 66.85it/s]




 48%|████▊     | 23826/50000 [06:29<06:27, 67.62it/s]




 48%|████▊     | 23833/50000 [06:29<06:26, 67.78it/s]




 48%|████▊     | 23841/50000 [06:29<06:24, 68.01it/s]




 48%|████▊     | 23848/50000 [06:29<06:29, 67.07it/s]




 48%|████▊     | 23855/50000 [06:29<06:34, 66.25it/s]




 48%|████▊     | 23862/50000 [06:29<06:39, 65.49it/s]




 48%|████▊     | 23869/50000 [06:29<06:46, 64.22it/s]




 48%|████▊     | 23876/50000 [06:30<06:44, 64.51it/s]




 48%|████▊     | 23884/50000 [06:30<06:32, 66.49it/s]




 48%|████▊     | 23891/50000 [06:30<06:39, 65.30it/s]




 48%|████▊     | 23898/50000 [06:30<06:33, 66.31it/s]




 48%|████▊     | 23905/50000 [06:30<07:04, 61.41it/s]




 48%|████▊     | 23912/50000 [06:30<06:58, 62.28it/s]




 48%|████▊     | 23919/50000 [06:30<06:59, 62.23it/s]




 48%|████▊     | 23926/50000 [06:30<06:46, 64.06it/s]




 48%|████▊     | 23933/50000 [06:30<06:52, 63.13it/s]




 48%|████▊     | 23940/50000 [06:31<06:59, 62.15it/s]




 48%|████▊     | 23947/50000 [06:31<07:00, 61.97it/s]




 48%|████▊     | 23954/50000 [06:31<07:04, 61.42it/s]




 48%|████▊     | 23961/50000 [06:31<06:55, 62.62it/s]




 48%|████▊     | 23969/50000 [06:31<06:36, 65.62it/s]




 48%|████▊     | 23976/50000 [06:31<06:43, 64.52it/s]




 48%|████▊     | 23983/50000 [06:31<06:42, 64.57it/s]




 48%|████▊     | 23990/50000 [06:31<06:56, 62.45it/s]




 48%|████▊     | 23997/50000 [06:32<07:11, 60.26it/s]




 48%|████▊     | 24004/50000 [06:32<06:55, 62.59it/s]




 48%|████▊     | 24011/50000 [06:32<06:48, 63.64it/s]




 48%|████▊     | 24018/50000 [06:32<06:42, 64.57it/s]




 48%|████▊     | 24025/50000 [06:32<06:39, 65.05it/s]




 48%|████▊     | 24032/50000 [06:32<07:03, 61.29it/s]




 48%|████▊     | 24039/50000 [06:32<06:58, 62.02it/s]




 48%|████▊     | 24046/50000 [06:32<06:55, 62.44it/s]




 48%|████▊     | 24054/50000 [06:32<06:37, 65.25it/s]




 48%|████▊     | 24061/50000 [06:32<06:33, 65.90it/s]




 48%|████▊     | 24068/50000 [06:33<06:33, 65.82it/s]




 48%|████▊     | 24075/50000 [06:33<06:46, 63.78it/s]




 48%|████▊     | 24082/50000 [06:33<06:44, 64.14it/s]




 48%|████▊     | 24089/50000 [06:33<06:36, 65.29it/s]




 48%|████▊     | 24096/50000 [06:33<06:29, 66.50it/s]




 48%|████▊     | 24103/50000 [06:33<06:33, 65.85it/s]




 48%|████▊     | 24110/50000 [06:33<06:32, 65.96it/s]




 48%|████▊     | 24117/50000 [06:33<06:33, 65.85it/s]




 48%|████▊     | 24124/50000 [06:33<06:27, 66.71it/s]




 48%|████▊     | 24131/50000 [06:34<06:36, 65.26it/s]




 48%|████▊     | 24139/50000 [06:34<06:16, 68.65it/s]




 48%|████▊     | 24147/50000 [06:34<06:15, 68.80it/s]




 48%|████▊     | 24154/50000 [06:34<06:27, 66.65it/s]




 48%|████▊     | 24162/50000 [06:34<06:17, 68.40it/s]




 48%|████▊     | 24169/50000 [06:34<06:37, 65.00it/s]




 48%|████▊     | 24176/50000 [06:34<06:40, 64.51it/s]




 48%|████▊     | 24184/50000 [06:34<06:20, 67.82it/s]




 48%|████▊     | 24191/50000 [06:34<06:22, 67.43it/s]




 48%|████▊     | 24198/50000 [06:35<06:21, 67.55it/s]




 48%|████▊     | 24205/50000 [06:35<06:41, 64.20it/s]




 48%|████▊     | 24213/50000 [06:35<06:26, 66.75it/s]




 48%|████▊     | 24220/50000 [06:35<06:29, 66.21it/s]




 48%|████▊     | 24228/50000 [06:35<06:17, 68.27it/s]




 48%|████▊     | 24235/50000 [06:35<06:19, 67.83it/s]




 48%|████▊     | 24242/50000 [06:35<06:34, 65.27it/s]




 48%|████▊     | 24249/50000 [06:35<06:47, 63.25it/s]




 49%|████▊     | 24256/50000 [06:35<07:01, 61.10it/s]




 49%|████▊     | 24263/50000 [06:36<06:49, 62.88it/s]




 49%|████▊     | 24271/50000 [06:36<06:27, 66.41it/s]




 49%|████▊     | 24278/50000 [06:36<06:39, 64.31it/s]




 49%|████▊     | 24285/50000 [06:36<06:35, 64.96it/s]




 49%|████▊     | 24293/50000 [06:36<06:20, 67.50it/s]




 49%|████▊     | 24302/50000 [06:36<06:01, 71.12it/s]




 49%|████▊     | 24310/50000 [06:36<05:51, 73.02it/s]




 49%|████▊     | 24318/50000 [06:36<05:56, 72.01it/s]




 49%|████▊     | 24326/50000 [06:36<06:17, 68.05it/s]




 49%|████▊     | 24334/50000 [06:37<06:08, 69.62it/s]




 49%|████▊     | 24342/50000 [06:37<06:22, 67.04it/s]




 49%|████▊     | 24350/50000 [06:37<06:04, 70.33it/s]




 49%|████▊     | 24358/50000 [06:37<05:58, 71.47it/s]




 49%|████▊     | 24366/50000 [06:37<06:45, 63.23it/s]




 49%|████▊     | 24373/50000 [06:37<07:15, 58.91it/s]




 49%|████▉     | 24380/50000 [06:37<07:31, 56.78it/s]




 49%|████▉     | 24386/50000 [06:37<07:27, 57.26it/s]




 49%|████▉     | 24392/50000 [06:38<07:43, 55.22it/s]




 49%|████▉     | 24398/50000 [06:38<08:25, 50.61it/s]




 49%|████▉     | 24404/50000 [06:38<08:32, 49.97it/s]




 49%|████▉     | 24410/50000 [06:38<09:03, 47.08it/s]




 49%|████▉     | 24415/50000 [06:38<09:10, 46.49it/s]




 49%|████▉     | 24421/50000 [06:38<08:34, 49.70it/s]




 49%|████▉     | 24427/50000 [06:38<08:24, 50.65it/s]




 49%|████▉     | 24433/50000 [06:38<08:38, 49.29it/s]




 49%|████▉     | 24439/50000 [06:39<08:28, 50.29it/s]




 49%|████▉     | 24446/50000 [06:39<07:50, 54.33it/s]




 49%|████▉     | 24453/50000 [06:39<07:27, 57.05it/s]




 49%|████▉     | 24459/50000 [06:39<07:24, 57.45it/s]




 49%|████▉     | 24467/50000 [06:39<06:56, 61.28it/s]




 49%|████▉     | 24474/50000 [06:39<06:58, 61.05it/s]




 49%|████▉     | 24481/50000 [06:39<06:48, 62.51it/s]




 49%|████▉     | 24488/50000 [06:39<06:45, 62.89it/s]




 49%|████▉     | 24495/50000 [06:39<06:51, 61.99it/s]




 49%|████▉     | 24502/50000 [06:40<07:15, 58.61it/s]




 49%|████▉     | 24508/50000 [06:40<07:40, 55.32it/s]




 49%|████▉     | 24514/50000 [06:40<07:35, 55.93it/s]




 49%|████▉     | 24520/50000 [06:40<08:04, 52.63it/s]




 49%|████▉     | 24526/50000 [06:40<08:11, 51.86it/s]




 49%|████▉     | 24532/50000 [06:40<08:04, 52.60it/s]




 49%|████▉     | 24538/50000 [06:40<07:54, 53.69it/s]




 49%|████▉     | 24544/50000 [06:40<08:44, 48.50it/s]




 49%|████▉     | 24551/50000 [06:40<08:00, 52.97it/s]




 49%|████▉     | 24557/50000 [06:41<08:40, 48.91it/s]




 49%|████▉     | 24564/50000 [06:41<07:53, 53.68it/s]




 49%|████▉     | 24571/50000 [06:41<07:31, 56.36it/s]




 49%|████▉     | 24579/50000 [06:41<06:58, 60.76it/s]




 49%|████▉     | 24586/50000 [06:41<06:57, 60.84it/s]




 49%|████▉     | 24593/50000 [06:41<07:14, 58.46it/s]




 49%|████▉     | 24601/50000 [06:41<06:42, 63.04it/s]




 49%|████▉     | 24608/50000 [06:41<06:46, 62.42it/s]




 49%|████▉     | 24616/50000 [06:42<06:25, 65.88it/s]




 49%|████▉     | 24623/50000 [06:42<06:32, 64.64it/s]




 49%|████▉     | 24631/50000 [06:42<06:16, 67.43it/s]




 49%|████▉     | 24638/50000 [06:42<06:25, 65.74it/s]




 49%|████▉     | 24645/50000 [06:42<06:28, 65.34it/s]




 49%|████▉     | 24652/50000 [06:42<06:22, 66.25it/s]




 49%|████▉     | 24659/50000 [06:42<06:22, 66.33it/s]




 49%|████▉     | 24666/50000 [06:42<06:28, 65.27it/s]




 49%|████▉     | 24673/50000 [06:42<06:34, 64.20it/s]




 49%|████▉     | 24680/50000 [06:42<06:28, 65.15it/s]




 49%|████▉     | 24687/50000 [06:43<06:26, 65.47it/s]




 49%|████▉     | 24694/50000 [06:43<06:50, 61.71it/s]




 49%|████▉     | 24701/50000 [06:43<07:16, 58.00it/s]




 49%|████▉     | 24707/50000 [06:43<07:18, 57.62it/s]




 49%|████▉     | 24714/50000 [06:43<06:57, 60.57it/s]




 49%|████▉     | 24721/50000 [06:43<06:48, 61.83it/s]




 49%|████▉     | 24729/50000 [06:43<06:34, 64.01it/s]




 49%|████▉     | 24736/50000 [06:43<06:40, 63.11it/s]




 49%|████▉     | 24743/50000 [06:44<06:47, 61.97it/s]




 50%|████▉     | 24750/50000 [06:44<06:42, 62.68it/s]




 50%|████▉     | 24758/50000 [06:44<06:29, 64.77it/s]




 50%|████▉     | 24766/50000 [06:44<06:13, 67.62it/s]




 50%|████▉     | 24774/50000 [06:44<06:01, 69.85it/s]




 50%|████▉     | 24782/50000 [06:44<05:50, 71.89it/s]




 50%|████▉     | 24790/50000 [06:44<06:09, 68.15it/s]




 50%|████▉     | 24797/50000 [06:44<06:30, 64.57it/s]




 50%|████▉     | 24804/50000 [06:44<06:47, 61.79it/s]




 50%|████▉     | 24811/50000 [06:45<07:19, 57.34it/s]




 50%|████▉     | 24817/50000 [06:45<07:28, 56.20it/s]




 50%|████▉     | 24824/50000 [06:45<07:11, 58.35it/s]




 50%|████▉     | 24831/50000 [06:45<06:51, 61.22it/s]




 50%|████▉     | 24839/50000 [06:45<06:29, 64.55it/s]




 50%|████▉     | 24846/50000 [06:45<06:36, 63.45it/s]




 50%|████▉     | 24853/50000 [06:45<06:36, 63.39it/s]




 50%|████▉     | 24860/50000 [06:45<06:37, 63.17it/s]




 50%|████▉     | 24867/50000 [06:45<07:00, 59.72it/s]




 50%|████▉     | 24874/50000 [06:46<07:00, 59.80it/s]




 50%|████▉     | 24881/50000 [06:46<06:48, 61.43it/s]




 50%|████▉     | 24888/50000 [06:46<06:36, 63.30it/s]




 50%|████▉     | 24895/50000 [06:46<06:28, 64.68it/s]




 50%|████▉     | 24902/50000 [06:46<06:27, 64.77it/s]




 50%|████▉     | 24909/50000 [06:46<06:23, 65.38it/s]




 50%|████▉     | 24916/50000 [06:46<06:25, 65.08it/s]




 50%|████▉     | 24924/50000 [06:46<06:11, 67.59it/s]




 50%|████▉     | 24931/50000 [06:46<06:25, 65.02it/s]




 50%|████▉     | 24938/50000 [06:47<06:25, 65.01it/s]




 50%|████▉     | 24945/50000 [06:47<07:01, 59.44it/s]




 50%|████▉     | 24953/50000 [06:47<06:34, 63.53it/s]




 50%|████▉     | 24961/50000 [06:47<06:21, 65.66it/s]




 50%|████▉     | 24968/50000 [06:47<06:33, 63.67it/s]




 50%|████▉     | 24975/50000 [06:47<06:25, 64.95it/s]




 50%|████▉     | 24982/50000 [06:47<06:17, 66.34it/s]




 50%|████▉     | 24989/50000 [06:47<06:30, 64.12it/s]




 50%|████▉     | 24996/50000 [06:47<06:21, 65.46it/s]




 50%|█████     | 25003/50000 [06:48<06:34, 63.37it/s]




 50%|█████     | 25011/50000 [06:48<06:15, 66.62it/s]




 50%|█████     | 25019/50000 [06:48<06:01, 69.10it/s]




 50%|█████     | 25026/50000 [06:48<06:11, 67.23it/s]




 50%|█████     | 25033/50000 [06:48<06:16, 66.35it/s]




 50%|█████     | 25040/50000 [06:48<06:17, 66.12it/s]




 50%|█████     | 25047/50000 [06:48<06:40, 62.28it/s]




 50%|█████     | 25054/50000 [06:48<06:43, 61.90it/s]




 50%|█████     | 25061/50000 [06:48<06:29, 64.11it/s]




 50%|█████     | 25068/50000 [06:49<06:23, 65.09it/s]




 50%|█████     | 25076/50000 [06:49<06:04, 68.46it/s]




 50%|█████     | 25083/50000 [06:49<06:30, 63.88it/s]




 50%|█████     | 25090/50000 [06:49<06:53, 60.24it/s]




 50%|█████     | 25097/50000 [06:49<07:09, 57.93it/s]




 50%|█████     | 25103/50000 [06:49<07:12, 57.57it/s]




 50%|█████     | 25109/50000 [06:49<07:21, 56.36it/s]




 50%|█████     | 25115/50000 [06:49<08:13, 50.44it/s]




 50%|█████     | 25121/50000 [06:50<07:54, 52.45it/s]




 50%|█████     | 25127/50000 [06:50<08:06, 51.08it/s]




 50%|█████     | 25134/50000 [06:50<07:31, 55.10it/s]




 50%|█████     | 25140/50000 [06:50<07:43, 53.68it/s]




 50%|█████     | 25148/50000 [06:50<06:59, 59.20it/s]




 50%|█████     | 25157/50000 [06:50<06:23, 64.73it/s]




 50%|█████     | 25166/50000 [06:50<06:04, 68.17it/s]




 50%|█████     | 25174/50000 [06:50<06:00, 68.81it/s]




 50%|█████     | 25182/50000 [06:50<05:52, 70.36it/s]




 50%|█████     | 25190/50000 [06:51<05:41, 72.66it/s]




 50%|█████     | 25198/50000 [06:51<06:19, 65.44it/s]




 50%|█████     | 25207/50000 [06:51<05:57, 69.35it/s]




 50%|█████     | 25216/50000 [06:51<05:39, 73.07it/s]




 50%|█████     | 25225/50000 [06:51<05:25, 76.11it/s]




 50%|█████     | 25233/50000 [06:51<05:22, 76.86it/s]




 50%|█████     | 25242/50000 [06:51<05:16, 78.22it/s]




 50%|█████     | 25250/50000 [06:51<05:22, 76.72it/s]




 51%|█████     | 25258/50000 [06:51<05:19, 77.49it/s]




 51%|█████     | 25266/50000 [06:52<05:39, 72.85it/s]




 51%|█████     | 25274/50000 [06:52<06:40, 61.76it/s]




 51%|█████     | 25281/50000 [06:52<06:45, 60.90it/s]




 51%|█████     | 25288/50000 [06:52<06:43, 61.26it/s]




 51%|█████     | 25295/50000 [06:52<07:05, 58.00it/s]




 51%|█████     | 25301/50000 [06:52<07:49, 52.63it/s]




 51%|█████     | 25307/50000 [06:52<08:03, 51.06it/s]




 51%|█████     | 25314/50000 [06:52<07:31, 54.70it/s]




 51%|█████     | 25321/50000 [06:53<07:07, 57.71it/s]




 51%|█████     | 25329/50000 [06:53<06:35, 62.35it/s]




 51%|█████     | 25336/50000 [06:53<06:39, 61.78it/s]




 51%|█████     | 25343/50000 [06:53<06:51, 59.98it/s]




 51%|█████     | 25350/50000 [06:53<08:17, 49.52it/s]




 51%|█████     | 25356/50000 [06:53<08:57, 45.89it/s]




 51%|█████     | 25361/50000 [06:53<08:54, 46.11it/s]




 51%|█████     | 25366/50000 [06:53<08:59, 45.70it/s]




 51%|█████     | 25371/50000 [06:54<09:00, 45.54it/s]




 51%|█████     | 25376/50000 [06:54<08:48, 46.56it/s]




 51%|█████     | 25381/50000 [06:54<08:43, 47.04it/s]




 51%|█████     | 25386/50000 [06:54<09:10, 44.72it/s]




 51%|█████     | 25392/50000 [06:54<08:35, 47.76it/s]




 51%|█████     | 25397/50000 [06:54<08:51, 46.30it/s]




 51%|█████     | 25403/50000 [06:54<08:29, 48.24it/s]




 51%|█████     | 25409/50000 [06:54<08:01, 51.09it/s]




 51%|█████     | 25415/50000 [06:55<08:27, 48.47it/s]




 51%|█████     | 25420/50000 [06:55<08:36, 47.58it/s]




 51%|█████     | 25426/50000 [06:55<08:18, 49.27it/s]




 51%|█████     | 25432/50000 [06:55<07:53, 51.84it/s]




 51%|█████     | 25438/50000 [06:55<07:43, 52.94it/s]




 51%|█████     | 25447/50000 [06:55<07:00, 58.44it/s]




 51%|█████     | 25454/50000 [06:55<06:43, 60.89it/s]




 51%|█████     | 25463/50000 [06:55<06:09, 66.43it/s]




 51%|█████     | 25471/50000 [06:55<05:55, 68.96it/s]




 51%|█████     | 25479/50000 [06:55<05:51, 69.73it/s]




 51%|█████     | 25487/50000 [06:56<06:32, 62.41it/s]




 51%|█████     | 25494/50000 [06:56<07:10, 56.87it/s]




 51%|█████     | 25500/50000 [06:56<07:24, 55.17it/s]




 51%|█████     | 25506/50000 [06:56<07:53, 51.69it/s]




 51%|█████     | 25512/50000 [06:56<07:57, 51.26it/s]




 51%|█████     | 25518/50000 [06:56<07:37, 53.49it/s]




 51%|█████     | 25524/50000 [06:56<07:40, 53.14it/s]




 51%|█████     | 25532/50000 [06:56<07:05, 57.47it/s]




 51%|█████     | 25540/50000 [06:57<06:36, 61.76it/s]




 51%|█████     | 25547/50000 [06:57<06:25, 63.46it/s]




 51%|█████     | 25554/50000 [06:57<06:17, 64.70it/s]




 51%|█████     | 25561/50000 [06:57<06:11, 65.83it/s]




 51%|█████     | 25570/50000 [06:57<05:46, 70.55it/s]




 51%|█████     | 25578/50000 [06:57<05:44, 70.98it/s]




 51%|█████     | 25586/50000 [06:57<06:15, 64.95it/s]




 51%|█████     | 25593/50000 [06:57<06:40, 60.90it/s]




 51%|█████     | 25602/50000 [06:58<06:08, 66.29it/s]




 51%|█████     | 25610/50000 [06:58<05:53, 69.03it/s]




 51%|█████     | 25618/50000 [06:58<06:35, 61.59it/s]




 51%|█████▏    | 25625/50000 [06:58<06:46, 60.00it/s]




 51%|█████▏    | 25632/50000 [06:58<06:51, 59.24it/s]




 51%|█████▏    | 25639/50000 [06:58<06:53, 58.86it/s]




 51%|█████▏    | 25645/50000 [06:58<06:53, 58.90it/s]




 51%|█████▏    | 25651/50000 [06:58<07:12, 56.28it/s]




 51%|█████▏    | 25657/50000 [06:58<07:23, 54.95it/s]




 51%|█████▏    | 25665/50000 [06:59<06:46, 59.93it/s]




 51%|█████▏    | 25673/50000 [06:59<06:17, 64.38it/s]




 51%|█████▏    | 25680/50000 [06:59<06:30, 62.33it/s]




 51%|█████▏    | 25687/50000 [06:59<06:32, 61.93it/s]




 51%|█████▏    | 25694/50000 [06:59<06:23, 63.33it/s]




 51%|█████▏    | 25703/50000 [06:59<06:01, 67.22it/s]




 51%|█████▏    | 25713/50000 [06:59<05:31, 73.29it/s]




 51%|█████▏    | 25721/50000 [06:59<05:29, 73.78it/s]




 51%|█████▏    | 25729/50000 [06:59<05:24, 74.86it/s]




 51%|█████▏    | 25737/50000 [07:00<05:33, 72.65it/s]




 51%|█████▏    | 25745/50000 [07:00<06:05, 66.41it/s]




 52%|█████▏    | 25752/50000 [07:00<06:14, 64.70it/s]




 52%|█████▏    | 25759/50000 [07:00<07:00, 57.61it/s]




 52%|█████▏    | 25765/50000 [07:00<07:06, 56.86it/s]




 52%|█████▏    | 25774/50000 [07:00<06:28, 62.37it/s]




 52%|█████▏    | 25781/50000 [07:00<06:54, 58.40it/s]




 52%|█████▏    | 25788/50000 [07:00<07:04, 57.01it/s]




 52%|█████▏    | 25796/50000 [07:01<06:34, 61.28it/s]




 52%|█████▏    | 25803/50000 [07:01<06:35, 61.21it/s]




 52%|█████▏    | 25810/50000 [07:01<06:45, 59.60it/s]




 52%|█████▏    | 25817/50000 [07:01<06:31, 61.84it/s]




 52%|█████▏    | 25826/50000 [07:01<06:03, 66.48it/s]




 52%|█████▏    | 25834/50000 [07:01<05:50, 68.88it/s]




 52%|█████▏    | 25842/50000 [07:01<06:18, 63.89it/s]




 52%|█████▏    | 25849/50000 [07:01<06:46, 59.38it/s]




 52%|█████▏    | 25856/50000 [07:02<06:46, 59.32it/s]




 52%|█████▏    | 25863/50000 [07:02<06:59, 57.54it/s]




 52%|█████▏    | 25870/50000 [07:02<06:45, 59.55it/s]




 52%|█████▏    | 25877/50000 [07:02<06:53, 58.27it/s]




 52%|█████▏    | 25884/50000 [07:02<06:54, 58.19it/s]




 52%|█████▏    | 25890/50000 [07:02<06:56, 57.92it/s]




 52%|█████▏    | 25896/50000 [07:02<07:11, 55.80it/s]




 52%|█████▏    | 25902/50000 [07:02<07:20, 54.71it/s]




 52%|█████▏    | 25909/50000 [07:02<07:06, 56.46it/s]




 52%|█████▏    | 25915/50000 [07:03<07:33, 53.11it/s]




 52%|█████▏    | 25921/50000 [07:03<07:33, 53.15it/s]




 52%|█████▏    | 25927/50000 [07:03<07:26, 53.96it/s]




 52%|█████▏    | 25934/50000 [07:03<07:11, 55.76it/s]




 52%|█████▏    | 25940/50000 [07:03<07:09, 56.05it/s]




 52%|█████▏    | 25946/50000 [07:03<07:23, 54.28it/s]




 52%|█████▏    | 25952/50000 [07:03<07:17, 55.00it/s]




 52%|█████▏    | 25959/50000 [07:03<07:04, 56.59it/s]




 52%|█████▏    | 25965/50000 [07:03<06:58, 57.37it/s]




 52%|█████▏    | 25972/50000 [07:04<06:40, 60.07it/s]




 52%|█████▏    | 25979/50000 [07:04<07:03, 56.70it/s]




 52%|█████▏    | 25988/50000 [07:04<06:21, 63.01it/s]




 52%|█████▏    | 25996/50000 [07:04<05:57, 67.18it/s]




 52%|█████▏    | 26003/50000 [07:04<06:26, 62.09it/s]




 52%|█████▏    | 26010/50000 [07:04<06:34, 60.81it/s]




 52%|█████▏    | 26017/50000 [07:04<06:46, 59.03it/s]




 52%|█████▏    | 26024/50000 [07:04<06:48, 58.72it/s]




 52%|█████▏    | 26030/50000 [07:05<06:46, 58.97it/s]




 52%|█████▏    | 26037/50000 [07:05<06:39, 60.03it/s]




 52%|█████▏    | 26044/50000 [07:05<06:50, 58.37it/s]




 52%|█████▏    | 26051/50000 [07:05<06:33, 60.82it/s]




 52%|█████▏    | 26058/50000 [07:05<06:41, 59.69it/s]




 52%|█████▏    | 26065/50000 [07:05<06:48, 58.58it/s]




 52%|█████▏    | 26072/50000 [07:05<06:34, 60.68it/s]




 52%|█████▏    | 26079/50000 [07:05<06:35, 60.47it/s]




 52%|█████▏    | 26086/50000 [07:05<06:58, 57.09it/s]




 52%|█████▏    | 26092/50000 [07:06<06:53, 57.81it/s]




 52%|█████▏    | 26098/50000 [07:06<07:11, 55.42it/s]




 52%|█████▏    | 26105/50000 [07:06<07:07, 55.91it/s]




 52%|█████▏    | 26111/50000 [07:06<07:01, 56.63it/s]




 52%|█████▏    | 26117/50000 [07:06<07:01, 56.67it/s]




 52%|█████▏    | 26123/50000 [07:06<07:15, 54.86it/s]




 52%|█████▏    | 26129/50000 [07:06<07:07, 55.88it/s]




 52%|█████▏    | 26135/50000 [07:06<07:17, 54.61it/s]




 52%|█████▏    | 26141/50000 [07:06<07:20, 54.12it/s]




 52%|█████▏    | 26149/50000 [07:07<06:49, 58.21it/s]




 52%|█████▏    | 26155/50000 [07:07<06:46, 58.70it/s]




 52%|█████▏    | 26162/50000 [07:07<06:39, 59.68it/s]




 52%|█████▏    | 26170/50000 [07:07<06:17, 63.12it/s]




 52%|█████▏    | 26178/50000 [07:07<06:03, 65.45it/s]




 52%|█████▏    | 26185/50000 [07:07<06:09, 64.41it/s]




 52%|█████▏    | 26192/50000 [07:07<06:05, 65.12it/s]




 52%|█████▏    | 26200/50000 [07:07<05:51, 67.79it/s]




 52%|█████▏    | 26207/50000 [07:07<05:55, 66.93it/s]




 52%|█████▏    | 26216/50000 [07:08<05:31, 71.69it/s]




 52%|█████▏    | 26224/50000 [07:08<05:27, 72.57it/s]




 52%|█████▏    | 26232/50000 [07:08<05:26, 72.90it/s]




 52%|█████▏    | 26240/50000 [07:08<05:34, 70.97it/s]




 52%|█████▏    | 26249/50000 [07:08<05:15, 75.26it/s]




 53%|█████▎    | 26258/50000 [07:08<05:10, 76.58it/s]




 53%|█████▎    | 26267/50000 [07:08<05:01, 78.65it/s]




 53%|█████▎    | 26275/50000 [07:08<05:02, 78.43it/s]




 53%|█████▎    | 26283/50000 [07:08<05:13, 75.71it/s]




 53%|█████▎    | 26292/50000 [07:09<05:06, 77.40it/s]




 53%|█████▎    | 26300/50000 [07:09<05:09, 76.54it/s]




 53%|█████▎    | 26309/50000 [07:09<05:01, 78.52it/s]




 53%|█████▎    | 26319/50000 [07:09<04:45, 82.94it/s]




 53%|█████▎    | 26328/50000 [07:09<05:17, 74.67it/s]




 53%|█████▎    | 26336/50000 [07:09<05:33, 70.99it/s]




 53%|█████▎    | 26344/50000 [07:09<05:43, 68.79it/s]




 53%|█████▎    | 26352/50000 [07:09<05:37, 69.98it/s]




 53%|█████▎    | 26360/50000 [07:10<05:51, 67.27it/s]




 53%|█████▎    | 26367/50000 [07:10<05:53, 66.76it/s]




 53%|█████▎    | 26374/50000 [07:10<05:55, 66.41it/s]




 53%|█████▎    | 26381/50000 [07:10<05:57, 66.06it/s]




 53%|█████▎    | 26388/50000 [07:10<06:08, 64.10it/s]




 53%|█████▎    | 26396/50000 [07:10<05:52, 66.93it/s]




 53%|█████▎    | 26403/50000 [07:10<05:51, 67.10it/s]




 53%|█████▎    | 26410/50000 [07:10<06:16, 62.71it/s]




 53%|█████▎    | 26417/50000 [07:10<06:12, 63.38it/s]




 53%|█████▎    | 26425/50000 [07:11<05:57, 65.97it/s]




 53%|█████▎    | 26436/50000 [07:11<05:21, 73.22it/s]




 53%|█████▎    | 26444/50000 [07:11<05:14, 74.97it/s]




 53%|█████▎    | 26454/50000 [07:11<04:57, 79.19it/s]




 53%|█████▎    | 26463/50000 [07:11<05:02, 77.75it/s]




 53%|█████▎    | 26471/50000 [07:11<05:13, 74.95it/s]




 53%|█████▎    | 26480/50000 [07:11<05:00, 78.14it/s]




 53%|█████▎    | 26488/50000 [07:11<05:37, 69.59it/s]




 53%|█████▎    | 26496/50000 [07:11<05:31, 70.81it/s]




 53%|█████▎    | 26504/50000 [07:12<05:35, 69.94it/s]




 53%|█████▎    | 26512/50000 [07:12<05:46, 67.76it/s]




 53%|█████▎    | 26519/50000 [07:12<06:00, 65.22it/s]




 53%|█████▎    | 26526/50000 [07:12<06:01, 64.97it/s]




 53%|█████▎    | 26533/50000 [07:12<06:05, 64.26it/s]




 53%|█████▎    | 26540/50000 [07:12<05:58, 65.38it/s]




 53%|█████▎    | 26548/50000 [07:12<05:43, 68.34it/s]




 53%|█████▎    | 26555/50000 [07:12<05:41, 68.68it/s]




 53%|█████▎    | 26565/50000 [07:12<05:15, 74.33it/s]




 53%|█████▎    | 26573/50000 [07:13<05:17, 73.72it/s]




 53%|█████▎    | 26581/50000 [07:13<05:21, 72.88it/s]




 53%|█████▎    | 26589/50000 [07:13<05:30, 70.77it/s]




 53%|█████▎    | 26597/50000 [07:13<05:43, 68.19it/s]




 53%|█████▎    | 26604/50000 [07:13<06:26, 60.58it/s]




 53%|█████▎    | 26611/50000 [07:13<06:29, 60.09it/s]




 53%|█████▎    | 26618/50000 [07:13<06:32, 59.60it/s]




 53%|█████▎    | 26626/50000 [07:13<06:04, 64.12it/s]




 53%|█████▎    | 26633/50000 [07:14<06:30, 59.77it/s]




 53%|█████▎    | 26640/50000 [07:14<06:56, 56.11it/s]




 53%|█████▎    | 26647/50000 [07:14<06:39, 58.50it/s]




 53%|█████▎    | 26655/50000 [07:14<06:10, 62.92it/s]




 53%|█████▎    | 26663/50000 [07:14<05:52, 66.27it/s]




 53%|█████▎    | 26670/50000 [07:14<05:59, 64.97it/s]




 53%|█████▎    | 26678/50000 [07:14<05:49, 66.66it/s]




 53%|█████▎    | 26685/50000 [07:14<06:09, 63.02it/s]




 53%|█████▎    | 26692/50000 [07:14<06:09, 63.00it/s]




 53%|█████▎    | 26699/50000 [07:15<06:05, 63.70it/s]




 53%|█████▎    | 26706/50000 [07:15<06:00, 64.61it/s]




 53%|█████▎    | 26715/50000 [07:15<05:35, 69.33it/s]




 53%|█████▎    | 26724/50000 [07:15<05:20, 72.53it/s]




 53%|█████▎    | 26732/50000 [07:15<05:28, 70.91it/s]




 53%|█████▎    | 26740/50000 [07:15<06:17, 61.62it/s]




 53%|█████▎    | 26747/50000 [07:15<06:18, 61.44it/s]




 54%|█████▎    | 26755/50000 [07:15<06:08, 63.12it/s]




 54%|█████▎    | 26762/50000 [07:16<06:48, 56.85it/s]




 54%|█████▎    | 26768/50000 [07:16<07:22, 52.51it/s]




 54%|█████▎    | 26774/50000 [07:16<07:59, 48.40it/s]




 54%|█████▎    | 26780/50000 [07:16<07:35, 51.02it/s]




 54%|█████▎    | 26790/50000 [07:16<06:28, 59.73it/s]




 54%|█████▎    | 26798/50000 [07:16<06:00, 64.37it/s]




 54%|█████▎    | 26807/50000 [07:16<05:45, 67.13it/s]




 54%|█████▎    | 26817/50000 [07:16<05:18, 72.89it/s]




 54%|█████▎    | 26825/50000 [07:16<05:30, 70.22it/s]




 54%|█████▎    | 26833/50000 [07:17<05:47, 66.75it/s]




 54%|█████▎    | 26840/50000 [07:17<06:17, 61.35it/s]




 54%|█████▎    | 26847/50000 [07:17<06:07, 63.07it/s]




 54%|█████▎    | 26854/50000 [07:17<06:00, 64.16it/s]




 54%|█████▎    | 26861/50000 [07:17<06:21, 60.73it/s]




 54%|█████▎    | 26868/50000 [07:17<06:20, 60.82it/s]




 54%|█████▍    | 26875/50000 [07:17<06:24, 60.10it/s]




 54%|█████▍    | 26882/50000 [07:17<06:11, 62.31it/s]




 54%|█████▍    | 26890/50000 [07:18<05:59, 64.37it/s]




 54%|█████▍    | 26897/50000 [07:18<06:26, 59.79it/s]




 54%|█████▍    | 26904/50000 [07:18<06:53, 55.85it/s]




 54%|█████▍    | 26911/50000 [07:18<06:39, 57.74it/s]




 54%|█████▍    | 26919/50000 [07:18<06:14, 61.58it/s]




 54%|█████▍    | 26926/50000 [07:18<06:19, 60.78it/s]




 54%|█████▍    | 26933/50000 [07:18<07:06, 54.10it/s]




 54%|█████▍    | 26940/50000 [07:18<06:47, 56.55it/s]




 54%|█████▍    | 26946/50000 [07:19<07:28, 51.39it/s]




 54%|█████▍    | 26952/50000 [07:19<07:10, 53.48it/s]




 54%|█████▍    | 26960/50000 [07:19<06:38, 57.75it/s]




 54%|█████▍    | 26966/50000 [07:19<06:34, 58.37it/s]




 54%|█████▍    | 26972/50000 [07:19<06:37, 57.96it/s]




 54%|█████▍    | 26980/50000 [07:19<06:14, 61.55it/s]




 54%|█████▍    | 26987/50000 [07:19<06:13, 61.63it/s]




 54%|█████▍    | 26994/50000 [07:19<06:16, 61.13it/s]




 54%|█████▍    | 27001/50000 [07:19<06:18, 60.78it/s]




 54%|█████▍    | 27008/50000 [07:20<06:10, 61.99it/s]




 54%|█████▍    | 27015/50000 [07:20<06:29, 59.05it/s]




 54%|█████▍    | 27022/50000 [07:20<06:15, 61.19it/s]




 54%|█████▍    | 27029/50000 [07:20<06:07, 62.45it/s]




 54%|█████▍    | 27036/50000 [07:20<06:34, 58.16it/s]




 54%|█████▍    | 27042/50000 [07:20<06:33, 58.32it/s]




 54%|█████▍    | 27049/50000 [07:20<06:30, 58.72it/s]




 54%|█████▍    | 27055/50000 [07:20<07:13, 52.91it/s]




 54%|█████▍    | 27062/50000 [07:21<06:43, 56.85it/s]




 54%|█████▍    | 27071/50000 [07:21<05:58, 63.87it/s]




 54%|█████▍    | 27078/50000 [07:21<06:09, 61.99it/s]




 54%|█████▍    | 27085/50000 [07:21<06:29, 58.85it/s]




 54%|█████▍    | 27093/50000 [07:21<06:08, 62.18it/s]




 54%|█████▍    | 27102/50000 [07:21<05:41, 67.05it/s]




 54%|█████▍    | 27110/50000 [07:21<05:34, 68.53it/s]




 54%|█████▍    | 27118/50000 [07:21<05:24, 70.53it/s]




 54%|█████▍    | 27126/50000 [07:21<05:26, 70.11it/s]




 54%|█████▍    | 27134/50000 [07:22<05:40, 67.19it/s]




 54%|█████▍    | 27141/50000 [07:22<06:06, 62.43it/s]




 54%|█████▍    | 27149/50000 [07:22<05:44, 66.38it/s]




 54%|█████▍    | 27156/50000 [07:22<05:59, 63.62it/s]




 54%|█████▍    | 27163/50000 [07:22<05:57, 63.85it/s]




 54%|█████▍    | 27171/50000 [07:22<05:43, 66.40it/s]




 54%|█████▍    | 27178/50000 [07:22<05:46, 65.78it/s]




 54%|█████▍    | 27185/50000 [07:22<06:03, 62.81it/s]




 54%|█████▍    | 27193/50000 [07:22<05:48, 65.38it/s]




 54%|█████▍    | 27200/50000 [07:23<06:49, 55.64it/s]




 54%|█████▍    | 27207/50000 [07:23<06:48, 55.79it/s]




 54%|█████▍    | 27213/50000 [07:23<07:04, 53.68it/s]




 54%|█████▍    | 27219/50000 [07:23<06:57, 54.56it/s]




 54%|█████▍    | 27226/50000 [07:23<06:38, 57.18it/s]




 54%|█████▍    | 27233/50000 [07:23<06:19, 59.93it/s]




 54%|█████▍    | 27240/50000 [07:23<06:32, 58.01it/s]




 54%|█████▍    | 27247/50000 [07:23<06:17, 60.25it/s]




 55%|█████▍    | 27254/50000 [07:24<06:19, 59.86it/s]




 55%|█████▍    | 27261/50000 [07:24<06:12, 60.99it/s]




 55%|█████▍    | 27268/50000 [07:24<05:59, 63.26it/s]




 55%|█████▍    | 27275/50000 [07:24<06:09, 61.50it/s]




 55%|█████▍    | 27282/50000 [07:24<06:19, 59.79it/s]




 55%|█████▍    | 27289/50000 [07:24<06:12, 60.94it/s]




 55%|█████▍    | 27296/50000 [07:24<06:02, 62.60it/s]




 55%|█████▍    | 27303/50000 [07:24<06:01, 62.79it/s]




 55%|█████▍    | 27310/50000 [07:24<06:12, 60.95it/s]




 55%|█████▍    | 27317/50000 [07:25<05:59, 63.12it/s]




 55%|█████▍    | 27324/50000 [07:25<06:08, 61.49it/s]




 55%|█████▍    | 27331/50000 [07:25<06:05, 62.00it/s]




 55%|█████▍    | 27339/50000 [07:25<05:45, 65.56it/s]




 55%|█████▍    | 27346/50000 [07:25<06:21, 59.41it/s]




 55%|█████▍    | 27353/50000 [07:25<06:48, 55.48it/s]




 55%|█████▍    | 27360/50000 [07:25<06:30, 57.95it/s]




 55%|█████▍    | 27366/50000 [07:25<06:56, 54.39it/s]




 55%|█████▍    | 27372/50000 [07:26<06:53, 54.70it/s]




 55%|█████▍    | 27378/50000 [07:26<07:12, 52.27it/s]




 55%|█████▍    | 27384/50000 [07:26<07:50, 48.03it/s]




 55%|█████▍    | 27389/50000 [07:26<07:52, 47.81it/s]




 55%|█████▍    | 27394/50000 [07:26<08:03, 46.72it/s]




 55%|█████▍    | 27399/50000 [07:26<09:20, 40.33it/s]




 55%|█████▍    | 27405/50000 [07:26<08:33, 43.97it/s]




 55%|█████▍    | 27410/50000 [07:26<09:25, 39.97it/s]




 55%|█████▍    | 27415/50000 [07:27<09:55, 37.92it/s]




 55%|█████▍    | 27420/50000 [07:27<09:18, 40.42it/s]




 55%|█████▍    | 27428/50000 [07:27<08:02, 46.82it/s]




 55%|█████▍    | 27436/50000 [07:27<07:16, 51.74it/s]




 55%|█████▍    | 27443/50000 [07:27<06:51, 54.85it/s]




 55%|█████▍    | 27449/50000 [07:27<06:58, 53.85it/s]




 55%|█████▍    | 27456/50000 [07:27<06:35, 57.04it/s]




 55%|█████▍    | 27463/50000 [07:27<06:16, 59.82it/s]




 55%|█████▍    | 27470/50000 [07:27<06:14, 60.18it/s]




 55%|█████▍    | 27477/50000 [07:28<06:14, 60.20it/s]




 55%|█████▍    | 27484/50000 [07:28<06:29, 57.76it/s]




 55%|█████▍    | 27491/50000 [07:28<06:19, 59.32it/s]




 55%|█████▍    | 27498/50000 [07:28<06:08, 61.00it/s]




 55%|█████▌    | 27505/50000 [07:28<06:28, 57.90it/s]




 55%|█████▌    | 27511/50000 [07:28<07:18, 51.29it/s]




 55%|█████▌    | 27517/50000 [07:28<07:43, 48.53it/s]




 55%|█████▌    | 27523/50000 [07:29<08:51, 42.29it/s]




 55%|█████▌    | 27528/50000 [07:29<08:37, 43.40it/s]




 55%|█████▌    | 27534/50000 [07:29<08:00, 46.75it/s]




 55%|█████▌    | 27541/50000 [07:29<07:22, 50.77it/s]




 55%|█████▌    | 27547/50000 [07:29<07:09, 52.29it/s]




 55%|█████▌    | 27553/50000 [07:29<07:04, 52.85it/s]




 55%|█████▌    | 27560/50000 [07:29<06:40, 55.99it/s]




 55%|█████▌    | 27566/50000 [07:29<06:42, 55.75it/s]




 55%|█████▌    | 27572/50000 [07:29<06:55, 53.93it/s]




 55%|█████▌    | 27578/50000 [07:30<07:29, 49.85it/s]




 55%|█████▌    | 27584/50000 [07:30<07:14, 51.61it/s]




 55%|█████▌    | 27591/50000 [07:30<06:53, 54.23it/s]




 55%|█████▌    | 27597/50000 [07:30<06:53, 54.13it/s]




 55%|█████▌    | 27603/50000 [07:30<07:02, 53.01it/s]




 55%|█████▌    | 27609/50000 [07:30<06:54, 54.08it/s]




 55%|█████▌    | 27615/50000 [07:30<06:50, 54.56it/s]




 55%|█████▌    | 27621/50000 [07:30<07:09, 52.11it/s]




 55%|█████▌    | 27629/50000 [07:30<06:33, 56.81it/s]




 55%|█████▌    | 27635/50000 [07:31<06:48, 54.78it/s]




 55%|█████▌    | 27641/50000 [07:31<06:58, 53.44it/s]




 55%|█████▌    | 27648/50000 [07:31<06:44, 55.25it/s]




 55%|█████▌    | 27655/50000 [07:31<06:24, 58.14it/s]




 55%|█████▌    | 27661/50000 [07:31<06:23, 58.22it/s]




 55%|█████▌    | 27668/50000 [07:31<06:11, 60.09it/s]




 55%|█████▌    | 27675/50000 [07:31<06:09, 60.37it/s]




 55%|█████▌    | 27682/50000 [07:31<06:14, 59.64it/s]




 55%|█████▌    | 27689/50000 [07:31<06:06, 60.83it/s]




 55%|█████▌    | 27696/50000 [07:32<06:04, 61.21it/s]




 55%|█████▌    | 27703/50000 [07:32<05:56, 62.63it/s]




 55%|█████▌    | 27710/50000 [07:32<06:17, 59.08it/s]




 55%|█████▌    | 27717/50000 [07:32<06:03, 61.29it/s]




 55%|█████▌    | 27724/50000 [07:32<06:02, 61.54it/s]




 55%|█████▌    | 27731/50000 [07:32<06:13, 59.66it/s]




 55%|█████▌    | 27738/50000 [07:32<06:21, 58.42it/s]




 55%|█████▌    | 27745/50000 [07:32<06:16, 59.18it/s]




 56%|█████▌    | 27752/50000 [07:33<06:02, 61.45it/s]




 56%|█████▌    | 27759/50000 [07:33<05:52, 63.14it/s]




 56%|█████▌    | 27766/50000 [07:33<05:48, 63.86it/s]




 56%|█████▌    | 27773/50000 [07:33<05:57, 62.23it/s]




 56%|█████▌    | 27780/50000 [07:33<05:59, 61.86it/s]




 56%|█████▌    | 27789/50000 [07:33<05:35, 66.20it/s]




 56%|█████▌    | 27796/50000 [07:33<05:35, 66.21it/s]




 56%|█████▌    | 27803/50000 [07:33<05:48, 63.69it/s]




 56%|█████▌    | 27810/50000 [07:33<06:01, 61.30it/s]




 56%|█████▌    | 27818/50000 [07:34<05:48, 63.69it/s]




 56%|█████▌    | 27825/50000 [07:34<06:06, 60.59it/s]




 56%|█████▌    | 27832/50000 [07:34<05:57, 62.01it/s]




 56%|█████▌    | 27839/50000 [07:34<05:54, 62.54it/s]




 56%|█████▌    | 27846/50000 [07:34<06:00, 61.43it/s]




 56%|█████▌    | 27853/50000 [07:34<05:47, 63.64it/s]




 56%|█████▌    | 27860/50000 [07:34<05:44, 64.21it/s]




 56%|█████▌    | 27867/50000 [07:34<05:44, 64.27it/s]




 56%|█████▌    | 27874/50000 [07:34<05:40, 65.02it/s]




 56%|█████▌    | 27881/50000 [07:35<06:02, 61.00it/s]




 56%|█████▌    | 27888/50000 [07:35<06:10, 59.76it/s]




 56%|█████▌    | 27895/50000 [07:35<05:57, 61.89it/s]




 56%|█████▌    | 27902/50000 [07:35<06:07, 60.21it/s]




 56%|█████▌    | 27909/50000 [07:35<06:04, 60.69it/s]




 56%|█████▌    | 27917/50000 [07:35<05:49, 63.15it/s]




 56%|█████▌    | 27924/50000 [07:35<05:44, 64.04it/s]




 56%|█████▌    | 27932/50000 [07:35<05:32, 66.47it/s]




 56%|█████▌    | 27939/50000 [07:35<05:41, 64.56it/s]




 56%|█████▌    | 27946/50000 [07:36<05:48, 63.29it/s]




 56%|█████▌    | 27953/50000 [07:36<06:13, 58.97it/s]




 56%|█████▌    | 27959/50000 [07:36<06:24, 57.29it/s]




 56%|█████▌    | 27966/50000 [07:36<06:05, 60.28it/s]




 56%|█████▌    | 27973/50000 [07:36<05:52, 62.52it/s]




 56%|█████▌    | 27980/50000 [07:36<05:56, 61.74it/s]




 56%|█████▌    | 27987/50000 [07:36<05:54, 62.18it/s]




 56%|█████▌    | 27994/50000 [07:36<05:44, 63.86it/s]




 56%|█████▌    | 28001/50000 [07:36<05:40, 64.55it/s]




 56%|█████▌    | 28008/50000 [07:37<06:00, 60.97it/s]




 56%|█████▌    | 28015/50000 [07:37<05:55, 61.79it/s]




 56%|█████▌    | 28022/50000 [07:37<05:57, 61.40it/s]




 56%|█████▌    | 28029/50000 [07:37<05:58, 61.29it/s]




 56%|█████▌    | 28037/50000 [07:37<05:33, 65.80it/s]




 56%|█████▌    | 28044/50000 [07:37<05:48, 63.08it/s]




 56%|█████▌    | 28052/50000 [07:37<05:31, 66.23it/s]




 56%|█████▌    | 28059/50000 [07:37<05:34, 65.67it/s]




 56%|█████▌    | 28066/50000 [07:37<05:39, 64.56it/s]




 56%|█████▌    | 28073/50000 [07:38<05:43, 63.80it/s]




 56%|█████▌    | 28080/50000 [07:38<05:58, 61.09it/s]




 56%|█████▌    | 28087/50000 [07:38<05:55, 61.72it/s]




 56%|█████▌    | 28094/50000 [07:38<05:53, 62.00it/s]




 56%|█████▌    | 28101/50000 [07:38<06:13, 58.70it/s]




 56%|█████▌    | 28109/50000 [07:38<05:47, 63.08it/s]




 56%|█████▌    | 28116/50000 [07:38<05:53, 61.96it/s]




 56%|█████▌    | 28123/50000 [07:38<05:47, 63.00it/s]




 56%|█████▋    | 28130/50000 [07:39<05:49, 62.57it/s]




 56%|█████▋    | 28137/50000 [07:39<05:58, 60.97it/s]




 56%|█████▋    | 28144/50000 [07:39<06:01, 60.51it/s]




 56%|█████▋    | 28151/50000 [07:39<05:51, 62.12it/s]




 56%|█████▋    | 28158/50000 [07:39<05:57, 61.14it/s]




 56%|█████▋    | 28165/50000 [07:39<05:51, 62.08it/s]




 56%|█████▋    | 28172/50000 [07:39<05:57, 61.12it/s]




 56%|█████▋    | 28180/50000 [07:39<05:40, 64.14it/s]




 56%|█████▋    | 28187/50000 [07:39<05:37, 64.57it/s]




 56%|█████▋    | 28194/50000 [07:40<05:39, 64.25it/s]




 56%|█████▋    | 28201/50000 [07:40<05:36, 64.74it/s]




 56%|█████▋    | 28208/50000 [07:40<05:40, 64.01it/s]




 56%|█████▋    | 28215/50000 [07:40<05:56, 61.11it/s]




 56%|█████▋    | 28223/50000 [07:40<05:39, 64.23it/s]




 56%|█████▋    | 28230/50000 [07:40<05:46, 62.90it/s]




 56%|█████▋    | 28237/50000 [07:40<05:51, 61.83it/s]




 56%|█████▋    | 28244/50000 [07:40<05:48, 62.41it/s]




 57%|█████▋    | 28251/50000 [07:40<05:38, 64.20it/s]




 57%|█████▋    | 28258/50000 [07:41<05:34, 64.97it/s]




 57%|█████▋    | 28265/50000 [07:41<05:40, 63.82it/s]




 57%|█████▋    | 28272/50000 [07:41<05:59, 60.36it/s]




 57%|█████▋    | 28279/50000 [07:41<05:50, 62.01it/s]




 57%|█████▋    | 28286/50000 [07:41<07:07, 50.74it/s]




 57%|█████▋    | 28292/50000 [07:41<07:14, 49.99it/s]




 57%|█████▋    | 28299/50000 [07:41<06:49, 52.97it/s]




 57%|█████▋    | 28305/50000 [07:41<06:43, 53.83it/s]




 57%|█████▋    | 28311/50000 [07:42<06:40, 54.09it/s]




 57%|█████▋    | 28317/50000 [07:42<07:20, 49.21it/s]




 57%|█████▋    | 28323/50000 [07:42<07:18, 49.48it/s]




 57%|█████▋    | 28329/50000 [07:42<07:32, 47.90it/s]




 57%|█████▋    | 28334/50000 [07:42<07:41, 46.91it/s]




 57%|█████▋    | 28339/50000 [07:42<08:19, 43.36it/s]




 57%|█████▋    | 28344/50000 [07:42<08:19, 43.32it/s]




 57%|█████▋    | 28349/50000 [07:42<08:05, 44.63it/s]




 57%|█████▋    | 28354/50000 [07:43<08:07, 44.43it/s]




 57%|█████▋    | 28361/50000 [07:43<07:21, 48.99it/s]




 57%|█████▋    | 28368/50000 [07:43<06:59, 51.55it/s]




 57%|█████▋    | 28374/50000 [07:43<06:43, 53.62it/s]




 57%|█████▋    | 28382/50000 [07:43<06:06, 59.02it/s]




 57%|█████▋    | 28390/50000 [07:43<05:45, 62.60it/s]




 57%|█████▋    | 28398/50000 [07:43<05:23, 66.85it/s]




 57%|█████▋    | 28405/50000 [07:43<05:27, 65.90it/s]




 57%|█████▋    | 28412/50000 [07:43<05:32, 64.90it/s]




 57%|█████▋    | 28419/50000 [07:44<05:31, 65.11it/s]




 57%|█████▋    | 28426/50000 [07:44<05:45, 62.46it/s]




 57%|█████▋    | 28433/50000 [07:44<05:52, 61.17it/s]




 57%|█████▋    | 28440/50000 [07:44<05:45, 62.46it/s]




 57%|█████▋    | 28447/50000 [07:44<06:07, 58.61it/s]




 57%|█████▋    | 28453/50000 [07:44<07:18, 49.15it/s]




 57%|█████▋    | 28459/50000 [07:44<07:29, 47.90it/s]




 57%|█████▋    | 28465/50000 [07:44<07:21, 48.79it/s]




 57%|█████▋    | 28472/50000 [07:45<06:47, 52.85it/s]




 57%|█████▋    | 28478/50000 [07:45<06:48, 52.69it/s]




 57%|█████▋    | 28485/50000 [07:45<06:22, 56.19it/s]




 57%|█████▋    | 28491/50000 [07:45<06:24, 55.90it/s]




 57%|█████▋    | 28497/50000 [07:45<06:25, 55.84it/s]




 57%|█████▋    | 28504/50000 [07:45<06:17, 56.89it/s]




 57%|█████▋    | 28511/50000 [07:45<06:10, 58.07it/s]




 57%|█████▋    | 28517/50000 [07:45<06:26, 55.59it/s]




 57%|█████▋    | 28524/50000 [07:45<06:08, 58.26it/s]




 57%|█████▋    | 28531/50000 [07:46<05:56, 60.28it/s]




 57%|█████▋    | 28538/50000 [07:46<06:12, 57.67it/s]




 57%|█████▋    | 28545/50000 [07:46<05:58, 59.84it/s]




 57%|█████▋    | 28552/50000 [07:46<05:50, 61.14it/s]




 57%|█████▋    | 28559/50000 [07:46<05:49, 61.43it/s]




 57%|█████▋    | 28566/50000 [07:46<05:42, 62.62it/s]




 57%|█████▋    | 28574/50000 [07:46<05:24, 66.04it/s]




 57%|█████▋    | 28581/50000 [07:46<05:28, 65.26it/s]




 57%|█████▋    | 28589/50000 [07:46<05:23, 66.21it/s]




 57%|█████▋    | 28597/50000 [07:47<05:09, 69.15it/s]




 57%|█████▋    | 28604/50000 [07:47<05:15, 67.85it/s]




 57%|█████▋    | 28612/50000 [07:47<05:12, 68.41it/s]




 57%|█████▋    | 28619/50000 [07:47<05:17, 67.34it/s]




 57%|█████▋    | 28626/50000 [07:47<05:18, 67.19it/s]




 57%|█████▋    | 28633/50000 [07:47<05:19, 66.90it/s]




 57%|█████▋    | 28640/50000 [07:47<05:26, 65.39it/s]




 57%|█████▋    | 28648/50000 [07:47<05:17, 67.31it/s]




 57%|█████▋    | 28655/50000 [07:47<05:25, 65.48it/s]




 57%|█████▋    | 28662/50000 [07:48<05:20, 66.63it/s]




 57%|█████▋    | 28670/50000 [07:48<05:12, 68.22it/s]




 57%|█████▋    | 28677/50000 [07:48<05:16, 67.41it/s]




 57%|█████▋    | 28684/50000 [07:48<05:25, 65.45it/s]




 57%|█████▋    | 28691/50000 [07:48<05:22, 66.05it/s]




 57%|█████▋    | 28698/50000 [07:48<05:30, 64.46it/s]




 57%|█████▋    | 28705/50000 [07:48<05:23, 65.89it/s]




 57%|█████▋    | 28712/50000 [07:48<05:18, 66.93it/s]




 57%|█████▋    | 28719/50000 [07:48<05:24, 65.59it/s]




 57%|█████▋    | 28727/50000 [07:48<05:12, 67.98it/s]




 57%|█████▋    | 28734/50000 [07:49<05:36, 63.24it/s]




 57%|█████▋    | 28741/50000 [07:49<05:52, 60.25it/s]




 57%|█████▋    | 28748/50000 [07:49<05:57, 59.48it/s]




 58%|█████▊    | 28755/50000 [07:49<05:52, 60.25it/s]




 58%|█████▊    | 28762/50000 [07:49<05:51, 60.48it/s]




 58%|█████▊    | 28770/50000 [07:49<05:31, 64.03it/s]




 58%|█████▊    | 28777/50000 [07:49<05:35, 63.18it/s]




 58%|█████▊    | 28784/50000 [07:49<05:35, 63.28it/s]




 58%|█████▊    | 28791/50000 [07:50<05:33, 63.61it/s]




 58%|█████▊    | 28798/50000 [07:50<05:52, 60.07it/s]




 58%|█████▊    | 28805/50000 [07:50<05:43, 61.79it/s]




 58%|█████▊    | 28812/50000 [07:50<05:42, 61.89it/s]




 58%|█████▊    | 28819/50000 [07:50<05:57, 59.29it/s]




 58%|█████▊    | 28826/50000 [07:50<05:42, 61.85it/s]




 58%|█████▊    | 28833/50000 [07:50<05:32, 63.62it/s]




 58%|█████▊    | 28840/50000 [07:50<05:44, 61.34it/s]




 58%|█████▊    | 28847/50000 [07:50<05:37, 62.72it/s]




 58%|█████▊    | 28854/50000 [07:51<05:45, 61.23it/s]




 58%|█████▊    | 28861/50000 [07:51<06:00, 58.56it/s]




 58%|█████▊    | 28867/50000 [07:51<06:03, 58.17it/s]




 58%|█████▊    | 28874/50000 [07:51<05:52, 59.99it/s]




 58%|█████▊    | 28882/50000 [07:51<05:32, 63.51it/s]




 58%|█████▊    | 28891/50000 [07:51<05:12, 67.51it/s]




 58%|█████▊    | 28899/50000 [07:51<05:07, 68.69it/s]




 58%|█████▊    | 28906/50000 [07:51<05:20, 65.83it/s]




 58%|█████▊    | 28913/50000 [07:51<05:26, 64.49it/s]




 58%|█████▊    | 28920/50000 [07:52<05:40, 61.90it/s]




 58%|█████▊    | 28927/50000 [07:52<05:42, 61.48it/s]




 58%|█████▊    | 28934/50000 [07:52<05:52, 59.70it/s]




 58%|█████▊    | 28942/50000 [07:52<05:26, 64.46it/s]




 58%|█████▊    | 28949/50000 [07:52<05:49, 60.26it/s]




 58%|█████▊    | 28957/50000 [07:52<05:28, 64.03it/s]




 58%|█████▊    | 28964/50000 [07:52<05:32, 63.27it/s]




 58%|█████▊    | 28971/50000 [07:52<05:39, 61.97it/s]




 58%|█████▊    | 28978/50000 [07:53<05:33, 63.08it/s]




 58%|█████▊    | 28985/50000 [07:53<05:34, 62.87it/s]




 58%|█████▊    | 28992/50000 [07:53<05:40, 61.65it/s]




 58%|█████▊    | 28999/50000 [07:53<05:42, 61.30it/s]




 58%|█████▊    | 29006/50000 [07:53<05:45, 60.75it/s]




 58%|█████▊    | 29013/50000 [07:53<05:32, 63.13it/s]




 58%|█████▊    | 29020/50000 [07:53<05:49, 60.08it/s]




 58%|█████▊    | 29027/50000 [07:53<05:42, 61.31it/s]




 58%|█████▊    | 29034/50000 [07:53<05:37, 62.20it/s]




 58%|█████▊    | 29041/50000 [07:54<05:28, 63.87it/s]




 58%|█████▊    | 29048/50000 [07:54<05:20, 65.46it/s]




 58%|█████▊    | 29055/50000 [07:54<05:21, 65.22it/s]




 58%|█████▊    | 29062/50000 [07:54<05:20, 65.37it/s]




 58%|█████▊    | 29070/50000 [07:54<05:14, 66.62it/s]




 58%|█████▊    | 29077/50000 [07:54<05:23, 64.66it/s]




 58%|█████▊    | 29085/50000 [07:54<05:08, 67.78it/s]




 58%|█████▊    | 29092/50000 [07:54<05:08, 67.70it/s]




 58%|█████▊    | 29099/50000 [07:54<05:12, 66.86it/s]




 58%|█████▊    | 29107/50000 [07:54<05:04, 68.57it/s]




 58%|█████▊    | 29115/50000 [07:55<05:04, 68.57it/s]




 58%|█████▊    | 29122/50000 [07:55<05:26, 63.95it/s]




 58%|█████▊    | 29129/50000 [07:55<05:20, 65.15it/s]




 58%|█████▊    | 29137/50000 [07:55<05:07, 67.82it/s]




 58%|█████▊    | 29144/50000 [07:55<05:21, 64.81it/s]




 58%|█████▊    | 29151/50000 [07:55<05:33, 62.52it/s]




 58%|█████▊    | 29158/50000 [07:55<05:26, 63.76it/s]




 58%|█████▊    | 29165/50000 [07:55<05:28, 63.48it/s]




 58%|█████▊    | 29172/50000 [07:56<05:36, 61.89it/s]




 58%|█████▊    | 29179/50000 [07:56<05:41, 60.99it/s]




 58%|█████▊    | 29186/50000 [07:56<05:50, 59.45it/s]




 58%|█████▊    | 29192/50000 [07:56<06:06, 56.79it/s]




 58%|█████▊    | 29198/50000 [07:56<06:04, 57.10it/s]




 58%|█████▊    | 29205/50000 [07:56<05:45, 60.17it/s]




 58%|█████▊    | 29212/50000 [07:56<06:06, 56.76it/s]




 58%|█████▊    | 29220/50000 [07:56<05:41, 60.81it/s]




 58%|█████▊    | 29227/50000 [07:56<05:40, 61.03it/s]




 58%|█████▊    | 29234/50000 [07:57<05:39, 61.19it/s]




 58%|█████▊    | 29241/50000 [07:57<05:47, 59.74it/s]




 58%|█████▊    | 29249/50000 [07:57<05:23, 64.23it/s]




 59%|█████▊    | 29256/50000 [07:57<05:25, 63.75it/s]




 59%|█████▊    | 29263/50000 [07:57<05:49, 59.40it/s]




 59%|█████▊    | 29270/50000 [07:57<06:19, 54.63it/s]




 59%|█████▊    | 29276/50000 [07:57<06:13, 55.48it/s]




 59%|█████▊    | 29282/50000 [07:57<06:31, 52.91it/s]




 59%|█████▊    | 29288/50000 [07:58<07:19, 47.15it/s]




 59%|█████▊    | 29294/50000 [07:58<07:06, 48.59it/s]




 59%|█████▊    | 29300/50000 [07:58<06:51, 50.27it/s]




 59%|█████▊    | 29306/50000 [07:58<07:04, 48.76it/s]




 59%|█████▊    | 29311/50000 [07:58<07:14, 47.63it/s]




 59%|█████▊    | 29316/50000 [07:58<07:12, 47.80it/s]




 59%|█████▊    | 29321/50000 [07:58<07:21, 46.84it/s]




 59%|█████▊    | 29326/50000 [07:58<07:33, 45.57it/s]




 59%|█████▊    | 29331/50000 [07:58<07:26, 46.33it/s]




 59%|█████▊    | 29337/50000 [07:59<07:06, 48.50it/s]




 59%|█████▊    | 29343/50000 [07:59<06:42, 51.32it/s]




 59%|█████▊    | 29350/50000 [07:59<06:19, 54.39it/s]




 59%|█████▊    | 29357/50000 [07:59<05:57, 57.75it/s]




 59%|█████▊    | 29365/50000 [07:59<05:34, 61.60it/s]




 59%|█████▊    | 29373/50000 [07:59<05:20, 64.45it/s]




 59%|█████▉    | 29380/50000 [07:59<05:13, 65.70it/s]




 59%|█████▉    | 29387/50000 [07:59<05:13, 65.67it/s]




 59%|█████▉    | 29394/50000 [07:59<05:20, 64.21it/s]




 59%|█████▉    | 29401/50000 [08:00<05:32, 61.88it/s]




 59%|█████▉    | 29408/50000 [08:00<05:31, 62.20it/s]




 59%|█████▉    | 29415/50000 [08:00<05:49, 58.92it/s]




 59%|█████▉    | 29421/50000 [08:00<06:03, 56.61it/s]




 59%|█████▉    | 29427/50000 [08:00<06:45, 50.77it/s]




 59%|█████▉    | 29433/50000 [08:00<06:35, 51.94it/s]




 59%|█████▉    | 29439/50000 [08:00<07:03, 48.53it/s]




 59%|█████▉    | 29444/50000 [08:00<07:03, 48.50it/s]




 59%|█████▉    | 29450/50000 [08:01<06:44, 50.77it/s]




 59%|█████▉    | 29456/50000 [08:01<06:50, 50.07it/s]




 59%|█████▉    | 29463/50000 [08:01<06:19, 54.15it/s]




 59%|█████▉    | 29470/50000 [08:01<06:05, 56.18it/s]




 59%|█████▉    | 29478/50000 [08:01<05:35, 61.16it/s]




 59%|█████▉    | 29485/50000 [08:01<05:26, 62.76it/s]




 59%|█████▉    | 29492/50000 [08:01<05:26, 62.74it/s]




 59%|█████▉    | 29500/50000 [08:01<05:15, 65.00it/s]




 59%|█████▉    | 29507/50000 [08:01<05:13, 65.27it/s]




 59%|█████▉    | 29514/50000 [08:02<05:17, 64.44it/s]




 59%|█████▉    | 29521/50000 [08:02<05:29, 62.20it/s]




 59%|█████▉    | 29528/50000 [08:02<05:35, 61.04it/s]




 59%|█████▉    | 29536/50000 [08:02<05:19, 63.96it/s]




 59%|█████▉    | 29543/50000 [08:02<05:33, 61.39it/s]




 59%|█████▉    | 29550/50000 [08:02<05:38, 60.49it/s]




 59%|█████▉    | 29557/50000 [08:02<05:35, 60.97it/s]




 59%|█████▉    | 29566/50000 [08:02<05:08, 66.20it/s]




 59%|█████▉    | 29573/50000 [08:02<05:15, 64.74it/s]




 59%|█████▉    | 29580/50000 [08:03<05:22, 63.41it/s]




 59%|█████▉    | 29588/50000 [08:03<05:10, 65.67it/s]




 59%|█████▉    | 29595/50000 [08:03<05:09, 66.02it/s]




 59%|█████▉    | 29602/50000 [08:03<05:08, 66.08it/s]




 59%|█████▉    | 29609/50000 [08:03<05:16, 64.48it/s]




 59%|█████▉    | 29616/50000 [08:03<05:23, 63.07it/s]




 59%|█████▉    | 29624/50000 [08:03<05:04, 66.83it/s]




 59%|█████▉    | 29631/50000 [08:03<05:05, 66.64it/s]




 59%|█████▉    | 29639/50000 [08:03<04:55, 68.85it/s]




 59%|█████▉    | 29646/50000 [08:04<05:13, 65.01it/s]




 59%|█████▉    | 29653/50000 [08:04<05:19, 63.59it/s]




 59%|█████▉    | 29660/50000 [08:04<05:28, 61.97it/s]




 59%|█████▉    | 29667/50000 [08:04<05:40, 59.64it/s]




 59%|█████▉    | 29674/50000 [08:04<05:42, 59.29it/s]




 59%|█████▉    | 29680/50000 [08:04<05:44, 59.02it/s]




 59%|█████▉    | 29687/50000 [08:04<05:38, 60.07it/s]




 59%|█████▉    | 29694/50000 [08:04<05:38, 60.05it/s]




 59%|█████▉    | 29701/50000 [08:05<05:37, 60.10it/s]




 59%|█████▉    | 29708/50000 [08:05<05:54, 57.27it/s]




 59%|█████▉    | 29715/50000 [08:05<05:37, 60.15it/s]




 59%|█████▉    | 29722/50000 [08:05<05:27, 61.96it/s]




 59%|█████▉    | 29729/50000 [08:05<05:32, 61.04it/s]




 59%|█████▉    | 29736/50000 [08:05<05:31, 61.20it/s]




 59%|█████▉    | 29743/50000 [08:05<05:25, 62.29it/s]




 60%|█████▉    | 29750/50000 [08:05<05:20, 63.24it/s]




 60%|█████▉    | 29757/50000 [08:05<05:34, 60.46it/s]




 60%|█████▉    | 29764/50000 [08:06<05:25, 62.08it/s]




 60%|█████▉    | 29772/50000 [08:06<05:14, 64.35it/s]




 60%|█████▉    | 29779/50000 [08:06<05:18, 63.49it/s]




 60%|█████▉    | 29786/50000 [08:06<05:23, 62.40it/s]




 60%|█████▉    | 29793/50000 [08:06<05:41, 59.16it/s]




 60%|█████▉    | 29799/50000 [08:06<05:46, 58.25it/s]




 60%|█████▉    | 29805/50000 [08:06<05:58, 56.33it/s]




 60%|█████▉    | 29811/50000 [08:06<05:56, 56.62it/s]




 60%|█████▉    | 29819/50000 [08:06<05:27, 61.67it/s]




 60%|█████▉    | 29826/50000 [08:07<05:18, 63.31it/s]




 60%|█████▉    | 29834/50000 [08:07<05:02, 66.74it/s]




 60%|█████▉    | 29841/50000 [08:07<05:08, 65.28it/s]




 60%|█████▉    | 29848/50000 [08:07<05:03, 66.49it/s]




 60%|█████▉    | 29855/50000 [08:07<05:17, 63.52it/s]




 60%|█████▉    | 29862/50000 [08:07<05:13, 64.31it/s]




 60%|█████▉    | 29869/50000 [08:07<05:15, 63.80it/s]




 60%|█████▉    | 29876/50000 [08:07<05:25, 61.90it/s]




 60%|█████▉    | 29884/50000 [08:07<05:08, 65.11it/s]




 60%|█████▉    | 29891/50000 [08:08<05:22, 62.37it/s]




 60%|█████▉    | 29898/50000 [08:08<05:14, 63.82it/s]




 60%|█████▉    | 29905/50000 [08:08<05:18, 63.13it/s]




 60%|█████▉    | 29912/50000 [08:08<05:23, 62.05it/s]




 60%|█████▉    | 29919/50000 [08:08<05:35, 59.80it/s]




 60%|█████▉    | 29926/50000 [08:08<05:34, 59.93it/s]




 60%|█████▉    | 29934/50000 [08:08<05:13, 64.07it/s]




 60%|█████▉    | 29941/50000 [08:08<05:06, 65.42it/s]




 60%|█████▉    | 29949/50000 [08:08<04:55, 67.85it/s]




 60%|█████▉    | 29956/50000 [08:09<05:03, 66.02it/s]




 60%|█████▉    | 29963/50000 [08:09<05:05, 65.52it/s]




 60%|█████▉    | 29971/50000 [08:09<04:58, 67.07it/s]




 60%|█████▉    | 29979/50000 [08:09<04:51, 68.73it/s]




 60%|█████▉    | 29986/50000 [08:09<04:56, 67.56it/s]




 60%|█████▉    | 29993/50000 [08:09<05:06, 65.37it/s]




 60%|██████    | 30001/50000 [08:09<04:57, 67.29it/s]




 60%|██████    | 30008/50000 [08:09<05:12, 64.03it/s]




 60%|██████    | 30015/50000 [08:09<05:32, 60.18it/s]




 60%|██████    | 30022/50000 [08:10<05:27, 60.98it/s]




 60%|██████    | 30029/50000 [08:10<05:26, 61.08it/s]




 60%|██████    | 30036/50000 [08:10<05:33, 59.81it/s]




 60%|██████    | 30043/50000 [08:10<05:23, 61.76it/s]




 60%|██████    | 30051/50000 [08:10<05:05, 65.21it/s]




 60%|██████    | 30058/50000 [08:10<05:07, 64.78it/s]




 60%|██████    | 30065/50000 [08:10<05:15, 63.09it/s]




 60%|██████    | 30072/50000 [08:10<05:17, 62.80it/s]




 60%|██████    | 30079/50000 [08:10<05:13, 63.61it/s]




 60%|██████    | 30087/50000 [08:11<04:59, 66.47it/s]




 60%|██████    | 30094/50000 [08:11<05:01, 66.02it/s]




 60%|██████    | 30101/50000 [08:11<05:16, 62.96it/s]




 60%|██████    | 30108/50000 [08:11<05:18, 62.54it/s]




 60%|██████    | 30115/50000 [08:11<06:37, 50.05it/s]




 60%|██████    | 30122/50000 [08:11<06:05, 54.38it/s]




 60%|██████    | 30129/50000 [08:11<05:44, 57.75it/s]




 60%|██████    | 30137/50000 [08:11<05:28, 60.54it/s]




 60%|██████    | 30144/50000 [08:12<05:24, 61.24it/s]




 60%|██████    | 30151/50000 [08:12<05:45, 57.52it/s]




 60%|██████    | 30159/50000 [08:12<05:23, 61.41it/s]




 60%|██████    | 30166/50000 [08:12<05:26, 60.66it/s]




 60%|██████    | 30173/50000 [08:12<06:05, 54.28it/s]




 60%|██████    | 30179/50000 [08:12<06:08, 53.82it/s]




 60%|██████    | 30185/50000 [08:12<06:04, 54.37it/s]




 60%|██████    | 30191/50000 [08:12<06:21, 51.93it/s]




 60%|██████    | 30197/50000 [08:13<06:06, 54.01it/s]




 60%|██████    | 30204/50000 [08:13<05:42, 57.88it/s]




 60%|██████    | 30210/50000 [08:13<05:41, 57.95it/s]




 60%|██████    | 30217/50000 [08:13<05:36, 58.84it/s]




 60%|██████    | 30225/50000 [08:13<05:26, 60.53it/s]




 60%|██████    | 30232/50000 [08:13<05:33, 59.30it/s]




 60%|██████    | 30238/50000 [08:13<06:00, 54.83it/s]




 60%|██████    | 30244/50000 [08:13<06:29, 50.70it/s]




 60%|██████    | 30250/50000 [08:14<06:36, 49.84it/s]




 61%|██████    | 30256/50000 [08:14<06:58, 47.22it/s]




 61%|██████    | 30262/50000 [08:14<06:41, 49.12it/s]




 61%|██████    | 30268/50000 [08:14<06:39, 49.42it/s]




 61%|██████    | 30274/50000 [08:14<06:20, 51.81it/s]




 61%|██████    | 30280/50000 [08:14<06:56, 47.30it/s]




 61%|██████    | 30286/50000 [08:14<06:33, 50.04it/s]




 61%|██████    | 30292/50000 [08:14<06:39, 49.39it/s]




 61%|██████    | 30298/50000 [08:14<06:46, 48.41it/s]




 61%|██████    | 30304/50000 [08:15<06:24, 51.16it/s]




 61%|██████    | 30311/50000 [08:15<06:03, 54.14it/s]




 61%|██████    | 30317/50000 [08:15<06:22, 51.43it/s]




 61%|██████    | 30325/50000 [08:15<05:48, 56.42it/s]




 61%|██████    | 30331/50000 [08:15<05:53, 55.58it/s]




 61%|██████    | 30338/50000 [08:15<05:35, 58.54it/s]




 61%|██████    | 30346/50000 [08:15<05:15, 62.37it/s]




 61%|██████    | 30353/50000 [08:15<05:25, 60.36it/s]




 61%|██████    | 30360/50000 [08:16<05:52, 55.79it/s]




 61%|██████    | 30366/50000 [08:16<06:12, 52.76it/s]




 61%|██████    | 30373/50000 [08:16<05:48, 56.32it/s]




 61%|██████    | 30379/50000 [08:16<06:18, 51.85it/s]




 61%|██████    | 30385/50000 [08:16<07:07, 45.92it/s]




 61%|██████    | 30390/50000 [08:16<07:01, 46.58it/s]




 61%|██████    | 30395/50000 [08:16<07:17, 44.84it/s]




 61%|██████    | 30400/50000 [08:16<07:27, 43.76it/s]




 61%|██████    | 30406/50000 [08:17<06:54, 47.26it/s]




 61%|██████    | 30412/50000 [08:17<06:29, 50.26it/s]




 61%|██████    | 30419/50000 [08:17<06:00, 54.30it/s]




 61%|██████    | 30425/50000 [08:17<06:03, 53.83it/s]




 61%|██████    | 30432/50000 [08:17<05:40, 57.45it/s]




 61%|██████    | 30440/50000 [08:17<05:18, 61.50it/s]




 61%|██████    | 30447/50000 [08:17<05:13, 62.34it/s]




 61%|██████    | 30454/50000 [08:17<05:03, 64.33it/s]




 61%|██████    | 30461/50000 [08:17<04:56, 65.79it/s]




 61%|██████    | 30468/50000 [08:17<05:04, 64.11it/s]




 61%|██████    | 30475/50000 [08:18<05:13, 62.31it/s]




 61%|██████    | 30482/50000 [08:18<05:06, 63.78it/s]




 61%|██████    | 30489/50000 [08:18<05:13, 62.18it/s]




 61%|██████    | 30496/50000 [08:18<05:16, 61.67it/s]




 61%|██████    | 30503/50000 [08:18<05:17, 61.31it/s]




 61%|██████    | 30510/50000 [08:18<05:19, 61.07it/s]




 61%|██████    | 30517/50000 [08:18<05:17, 61.38it/s]




 61%|██████    | 30524/50000 [08:18<05:17, 61.27it/s]




 61%|██████    | 30531/50000 [08:19<05:13, 62.14it/s]




 61%|██████    | 30538/50000 [08:19<05:09, 62.96it/s]




 61%|██████    | 30546/50000 [08:19<04:57, 65.49it/s]




 61%|██████    | 30553/50000 [08:19<05:04, 63.91it/s]




 61%|██████    | 30560/50000 [08:19<05:16, 61.37it/s]




 61%|██████    | 30568/50000 [08:19<04:59, 64.90it/s]




 61%|██████    | 30575/50000 [08:19<04:57, 65.29it/s]




 61%|██████    | 30583/50000 [08:19<04:52, 66.40it/s]




 61%|██████    | 30590/50000 [08:19<04:55, 65.60it/s]




 61%|██████    | 30597/50000 [08:20<05:11, 62.28it/s]




 61%|██████    | 30604/50000 [08:20<05:05, 63.58it/s]




 61%|██████    | 30611/50000 [08:20<05:27, 59.14it/s]




 61%|██████    | 30618/50000 [08:20<05:43, 56.46it/s]




 61%|██████▏   | 30626/50000 [08:20<05:17, 61.11it/s]




 61%|██████▏   | 30634/50000 [08:20<05:04, 63.62it/s]




 61%|██████▏   | 30642/50000 [08:20<04:48, 67.15it/s]




 61%|██████▏   | 30650/50000 [08:20<04:39, 69.32it/s]




 61%|██████▏   | 30658/50000 [08:20<04:50, 66.51it/s]




 61%|██████▏   | 30665/50000 [08:21<04:52, 66.05it/s]




 61%|██████▏   | 30672/50000 [08:21<04:55, 65.36it/s]




 61%|██████▏   | 30680/50000 [08:21<04:50, 66.45it/s]




 61%|██████▏   | 30689/50000 [08:21<04:31, 71.22it/s]




 61%|██████▏   | 30697/50000 [08:21<04:39, 68.95it/s]




 61%|██████▏   | 30704/50000 [08:21<04:44, 67.71it/s]




 61%|██████▏   | 30711/50000 [08:21<05:03, 63.60it/s]




 61%|██████▏   | 30718/50000 [08:21<05:16, 61.00it/s]




 61%|██████▏   | 30725/50000 [08:22<05:13, 61.40it/s]




 61%|██████▏   | 30733/50000 [08:22<04:56, 64.89it/s]




 61%|██████▏   | 30741/50000 [08:22<04:45, 67.56it/s]




 62%|██████▏   | 30750/50000 [08:22<04:32, 70.67it/s]




 62%|██████▏   | 30758/50000 [08:22<04:36, 69.66it/s]




 62%|██████▏   | 30766/50000 [08:22<04:41, 68.44it/s]




 62%|██████▏   | 30773/50000 [08:22<04:46, 67.17it/s]




 62%|██████▏   | 30780/50000 [08:22<04:44, 67.65it/s]




 62%|██████▏   | 30787/50000 [08:22<04:49, 66.45it/s]




 62%|██████▏   | 30794/50000 [08:23<04:57, 64.55it/s]




 62%|██████▏   | 30802/50000 [08:23<04:50, 66.03it/s]




 62%|██████▏   | 30809/50000 [08:23<04:53, 65.41it/s]




 62%|██████▏   | 30817/50000 [08:23<04:42, 68.01it/s]




 62%|██████▏   | 30824/50000 [08:23<04:54, 65.02it/s]




 62%|██████▏   | 30831/50000 [08:23<04:54, 65.00it/s]




 62%|██████▏   | 30838/50000 [08:23<04:57, 64.37it/s]




 62%|██████▏   | 30845/50000 [08:23<04:55, 64.73it/s]




 62%|██████▏   | 30852/50000 [08:23<04:52, 65.35it/s]




 62%|██████▏   | 30859/50000 [08:23<04:49, 66.16it/s]




 62%|██████▏   | 30867/50000 [08:24<04:41, 67.88it/s]




 62%|██████▏   | 30875/50000 [08:24<04:34, 69.68it/s]




 62%|██████▏   | 30882/50000 [08:24<04:46, 66.65it/s]




 62%|██████▏   | 30889/50000 [08:24<04:53, 65.04it/s]




 62%|██████▏   | 30896/50000 [08:24<04:48, 66.13it/s]




 62%|██████▏   | 30903/50000 [08:24<04:50, 65.69it/s]




 62%|██████▏   | 30910/50000 [08:24<04:59, 63.69it/s]




 62%|██████▏   | 30917/50000 [08:24<04:57, 64.25it/s]




 62%|██████▏   | 30924/50000 [08:24<04:55, 64.47it/s]




 62%|██████▏   | 30931/50000 [08:25<04:59, 63.73it/s]




 62%|██████▏   | 30938/50000 [08:25<05:02, 62.98it/s]




 62%|██████▏   | 30946/50000 [08:25<04:44, 66.90it/s]




 62%|██████▏   | 30953/50000 [08:25<05:06, 62.09it/s]




 62%|██████▏   | 30960/50000 [08:25<04:58, 63.79it/s]




 62%|██████▏   | 30967/50000 [08:25<05:17, 59.88it/s]




 62%|██████▏   | 30974/50000 [08:25<05:15, 60.37it/s]




 62%|██████▏   | 30981/50000 [08:25<05:02, 62.85it/s]




 62%|██████▏   | 30988/50000 [08:26<05:02, 62.79it/s]




 62%|██████▏   | 30995/50000 [08:26<05:15, 60.33it/s]




 62%|██████▏   | 31003/50000 [08:26<04:57, 63.92it/s]




 62%|██████▏   | 31010/50000 [08:26<05:11, 61.03it/s]




 62%|██████▏   | 31017/50000 [08:26<05:04, 62.25it/s]




 62%|██████▏   | 31024/50000 [08:26<04:56, 64.08it/s]




 62%|██████▏   | 31031/50000 [08:26<04:58, 63.48it/s]




 62%|██████▏   | 31039/50000 [08:26<04:42, 67.21it/s]




 62%|██████▏   | 31046/50000 [08:26<04:55, 64.15it/s]




 62%|██████▏   | 31053/50000 [08:27<05:05, 61.94it/s]




 62%|██████▏   | 31062/50000 [08:27<04:45, 66.43it/s]




 62%|██████▏   | 31069/50000 [08:27<04:46, 66.18it/s]




 62%|██████▏   | 31076/50000 [08:27<04:49, 65.27it/s]




 62%|██████▏   | 31084/50000 [08:27<04:41, 67.22it/s]




 62%|██████▏   | 31091/50000 [08:27<04:49, 65.42it/s]




 62%|██████▏   | 31098/50000 [08:27<04:51, 64.75it/s]




 62%|██████▏   | 31106/50000 [08:27<04:39, 67.68it/s]




 62%|██████▏   | 31113/50000 [08:27<04:56, 63.75it/s]




 62%|██████▏   | 31120/50000 [08:28<05:00, 62.91it/s]




 62%|██████▏   | 31127/50000 [08:28<05:18, 59.33it/s]




 62%|██████▏   | 31135/50000 [08:28<04:59, 63.03it/s]




 62%|██████▏   | 31142/50000 [08:28<04:55, 63.78it/s]




 62%|██████▏   | 31150/50000 [08:28<04:42, 66.77it/s]




 62%|██████▏   | 31158/50000 [08:28<04:35, 68.50it/s]




 62%|██████▏   | 31165/50000 [08:28<04:55, 63.74it/s]




 62%|██████▏   | 31173/50000 [08:28<04:48, 65.27it/s]




 62%|██████▏   | 31180/50000 [08:28<04:55, 63.59it/s]




 62%|██████▏   | 31187/50000 [08:29<04:53, 64.18it/s]




 62%|██████▏   | 31195/50000 [08:29<04:40, 67.07it/s]




 62%|██████▏   | 31202/50000 [08:29<04:43, 66.25it/s]




 62%|██████▏   | 31209/50000 [08:29<04:56, 63.37it/s]




 62%|██████▏   | 31216/50000 [08:29<05:10, 60.46it/s]




 62%|██████▏   | 31223/50000 [08:29<05:20, 58.65it/s]




 62%|██████▏   | 31229/50000 [08:29<05:34, 56.12it/s]




 62%|██████▏   | 31235/50000 [08:29<06:02, 51.80it/s]




 62%|██████▏   | 31241/50000 [08:30<06:19, 49.39it/s]




 62%|██████▏   | 31247/50000 [08:30<06:29, 48.20it/s]




 63%|██████▎   | 31252/50000 [08:30<06:30, 47.99it/s]




 63%|██████▎   | 31257/50000 [08:30<06:58, 44.83it/s]




 63%|██████▎   | 31263/50000 [08:30<06:29, 48.08it/s]




 63%|██████▎   | 31268/50000 [08:30<06:41, 46.64it/s]




 63%|██████▎   | 31274/50000 [08:30<06:21, 49.03it/s]




 63%|██████▎   | 31281/50000 [08:30<05:59, 52.11it/s]




 63%|██████▎   | 31287/50000 [08:31<06:17, 49.59it/s]




 63%|██████▎   | 31293/50000 [08:31<06:01, 51.81it/s]




 63%|██████▎   | 31299/50000 [08:31<05:47, 53.82it/s]




 63%|██████▎   | 31306/50000 [08:31<05:29, 56.74it/s]




 63%|██████▎   | 31313/50000 [08:31<05:17, 58.84it/s]




 63%|██████▎   | 31319/50000 [08:31<05:20, 58.20it/s]




 63%|██████▎   | 31326/50000 [08:31<05:09, 60.28it/s]




 63%|██████▎   | 31333/50000 [08:31<05:06, 60.82it/s]




 63%|██████▎   | 31340/50000 [08:31<05:01, 61.93it/s]




 63%|██████▎   | 31348/50000 [08:31<04:45, 65.34it/s]




 63%|██████▎   | 31355/50000 [08:32<04:42, 65.97it/s]




 63%|██████▎   | 31362/50000 [08:32<05:16, 58.82it/s]




 63%|██████▎   | 31369/50000 [08:32<05:26, 56.98it/s]




 63%|██████▎   | 31375/50000 [08:32<05:54, 52.58it/s]




 63%|██████▎   | 31381/50000 [08:32<06:15, 49.64it/s]




 63%|██████▎   | 31387/50000 [08:32<06:11, 50.16it/s]




 63%|██████▎   | 31393/50000 [08:32<06:44, 45.96it/s]




 63%|██████▎   | 31399/50000 [08:33<06:32, 47.36it/s]




 63%|██████▎   | 31404/50000 [08:33<06:31, 47.47it/s]




 63%|██████▎   | 31411/50000 [08:33<05:59, 51.65it/s]




 63%|██████▎   | 31419/50000 [08:33<05:25, 57.02it/s]




 63%|██████▎   | 31426/50000 [08:33<05:10, 59.88it/s]




 63%|██████▎   | 31433/50000 [08:33<05:10, 59.76it/s]




 63%|██████▎   | 31440/50000 [08:33<05:07, 60.45it/s]




 63%|██████▎   | 31447/50000 [08:33<05:10, 59.69it/s]




 63%|██████▎   | 31454/50000 [08:33<05:12, 59.33it/s]




 63%|██████▎   | 31460/50000 [08:34<05:18, 58.19it/s]




 63%|██████▎   | 31468/50000 [08:34<04:58, 62.09it/s]




 63%|██████▎   | 31476/50000 [08:34<04:49, 64.05it/s]




 63%|██████▎   | 31483/50000 [08:34<04:49, 63.98it/s]




 63%|██████▎   | 31491/50000 [08:34<04:34, 67.43it/s]




 63%|██████▎   | 31498/50000 [08:34<04:35, 67.06it/s]




 63%|██████▎   | 31505/50000 [08:34<04:46, 64.59it/s]




 63%|██████▎   | 31513/50000 [08:34<04:39, 66.23it/s]




 63%|██████▎   | 31520/50000 [08:34<04:49, 63.79it/s]




 63%|██████▎   | 31527/50000 [08:35<04:52, 63.10it/s]




 63%|██████▎   | 31534/50000 [08:35<04:58, 61.81it/s]




 63%|██████▎   | 31541/50000 [08:35<04:54, 62.73it/s]




 63%|██████▎   | 31548/50000 [08:35<04:46, 64.44it/s]




 63%|██████▎   | 31555/50000 [08:35<05:01, 61.22it/s]




 63%|██████▎   | 31563/50000 [08:35<04:43, 65.10it/s]




 63%|██████▎   | 31571/50000 [08:35<04:36, 66.76it/s]




 63%|██████▎   | 31578/50000 [08:35<04:37, 66.41it/s]




 63%|██████▎   | 31585/50000 [08:35<05:01, 61.06it/s]




 63%|██████▎   | 31592/50000 [08:36<04:59, 61.37it/s]




 63%|██████▎   | 31599/50000 [08:36<04:53, 62.75it/s]




 63%|██████▎   | 31607/50000 [08:36<04:42, 65.01it/s]




 63%|██████▎   | 31614/50000 [08:36<04:50, 63.33it/s]




 63%|██████▎   | 31622/50000 [08:36<04:37, 66.26it/s]




 63%|██████▎   | 31630/50000 [08:36<04:24, 69.36it/s]




 63%|██████▎   | 31638/50000 [08:36<04:30, 67.89it/s]




 63%|██████▎   | 31645/50000 [08:36<04:43, 64.77it/s]




 63%|██████▎   | 31652/50000 [08:36<04:46, 63.95it/s]




 63%|██████▎   | 31660/50000 [08:37<04:32, 67.40it/s]




 63%|██████▎   | 31667/50000 [08:37<04:35, 66.66it/s]




 63%|██████▎   | 31674/50000 [08:37<04:36, 66.34it/s]




 63%|██████▎   | 31681/50000 [08:37<04:41, 65.10it/s]




 63%|██████▎   | 31688/50000 [08:37<04:58, 61.31it/s]




 63%|██████▎   | 31695/50000 [08:37<05:04, 60.04it/s]




 63%|██████▎   | 31703/50000 [08:37<04:46, 63.94it/s]




 63%|██████▎   | 31711/50000 [08:37<04:34, 66.64it/s]




 63%|██████▎   | 31718/50000 [08:37<04:33, 66.82it/s]




 63%|██████▎   | 31725/50000 [08:38<04:44, 64.17it/s]




 63%|██████▎   | 31733/50000 [08:38<04:32, 66.98it/s]




 63%|██████▎   | 31740/50000 [08:38<04:40, 65.08it/s]




 63%|██████▎   | 31747/50000 [08:38<04:40, 65.05it/s]




 64%|██████▎   | 31754/50000 [08:38<04:39, 65.21it/s]




 64%|██████▎   | 31762/50000 [08:38<04:31, 67.18it/s]




 64%|██████▎   | 31769/50000 [08:38<04:39, 65.21it/s]




 64%|██████▎   | 31776/50000 [08:38<04:44, 64.07it/s]




 64%|██████▎   | 31783/50000 [08:38<04:46, 63.64it/s]




 64%|██████▎   | 31790/50000 [08:39<04:56, 61.35it/s]




 64%|██████▎   | 31798/50000 [08:39<04:39, 65.12it/s]




 64%|██████▎   | 31805/50000 [08:39<04:47, 63.34it/s]




 64%|██████▎   | 31812/50000 [08:39<04:51, 62.46it/s]




 64%|██████▎   | 31820/50000 [08:39<04:37, 65.50it/s]




 64%|██████▎   | 31827/50000 [08:39<04:37, 65.53it/s]




 64%|██████▎   | 31835/50000 [08:39<04:27, 68.01it/s]




 64%|██████▎   | 31842/50000 [08:39<04:47, 63.27it/s]




 64%|██████▎   | 31849/50000 [08:39<04:46, 63.26it/s]




 64%|██████▎   | 31856/50000 [08:40<04:56, 61.26it/s]




 64%|██████▎   | 31863/50000 [08:40<04:57, 60.87it/s]




 64%|██████▎   | 31871/50000 [08:40<04:44, 63.74it/s]




 64%|██████▍   | 31879/50000 [08:40<04:30, 67.08it/s]




 64%|██████▍   | 31886/50000 [08:40<04:29, 67.20it/s]




 64%|██████▍   | 31893/50000 [08:40<04:48, 62.85it/s]




 64%|██████▍   | 31900/50000 [08:40<04:59, 60.52it/s]




 64%|██████▍   | 31907/50000 [08:40<05:09, 58.40it/s]




 64%|██████▍   | 31914/50000 [08:41<05:04, 59.32it/s]




 64%|██████▍   | 31921/50000 [08:41<04:53, 61.63it/s]




 64%|██████▍   | 31928/50000 [08:41<04:59, 60.42it/s]




 64%|██████▍   | 31935/50000 [08:41<04:58, 60.44it/s]




 64%|██████▍   | 31942/50000 [08:41<05:04, 59.39it/s]




 64%|██████▍   | 31949/50000 [08:41<04:51, 61.93it/s]




 64%|██████▍   | 31956/50000 [08:41<04:45, 63.16it/s]




 64%|██████▍   | 31963/50000 [08:41<05:06, 58.89it/s]




 64%|██████▍   | 31970/50000 [08:41<05:03, 59.36it/s]




 64%|██████▍   | 31976/50000 [08:42<05:04, 59.25it/s]




 64%|██████▍   | 31984/50000 [08:42<04:47, 62.63it/s]




 64%|██████▍   | 31992/50000 [08:42<04:29, 66.71it/s]




 64%|██████▍   | 31999/50000 [08:42<04:45, 63.14it/s]




 64%|██████▍   | 32006/50000 [08:42<04:42, 63.69it/s]




 64%|██████▍   | 32013/50000 [08:42<04:37, 64.83it/s]




 64%|██████▍   | 32021/50000 [08:42<04:28, 66.90it/s]




 64%|██████▍   | 32028/50000 [08:42<04:55, 60.79it/s]




 64%|██████▍   | 32035/50000 [08:42<04:47, 62.49it/s]




 64%|██████▍   | 32042/50000 [08:43<04:51, 61.56it/s]




 64%|██████▍   | 32049/50000 [08:43<04:49, 62.05it/s]




 64%|██████▍   | 32056/50000 [08:43<04:49, 62.07it/s]




 64%|██████▍   | 32063/50000 [08:43<04:48, 62.08it/s]




 64%|██████▍   | 32070/50000 [08:43<05:06, 58.52it/s]




 64%|██████▍   | 32078/50000 [08:43<04:48, 62.06it/s]




 64%|██████▍   | 32085/50000 [08:43<04:48, 61.99it/s]




 64%|██████▍   | 32092/50000 [08:43<04:44, 62.86it/s]




 64%|██████▍   | 32099/50000 [08:44<04:48, 62.06it/s]




 64%|██████▍   | 32106/50000 [08:44<04:49, 61.91it/s]




 64%|██████▍   | 32113/50000 [08:44<04:39, 64.00it/s]




 64%|██████▍   | 32120/50000 [08:44<04:35, 64.98it/s]




 64%|██████▍   | 32127/50000 [08:44<04:40, 63.74it/s]




 64%|██████▍   | 32134/50000 [08:44<04:41, 63.42it/s]




 64%|██████▍   | 32141/50000 [08:44<04:51, 61.21it/s]




 64%|██████▍   | 32148/50000 [08:44<04:41, 63.31it/s]




 64%|██████▍   | 32155/50000 [08:44<04:48, 61.78it/s]




 64%|██████▍   | 32163/50000 [08:44<04:32, 65.38it/s]




 64%|██████▍   | 32170/50000 [08:45<04:41, 63.32it/s]




 64%|██████▍   | 32178/50000 [08:45<04:30, 65.92it/s]




 64%|██████▍   | 32185/50000 [08:45<04:44, 62.65it/s]




 64%|██████▍   | 32192/50000 [08:45<04:37, 64.20it/s]




 64%|██████▍   | 32199/50000 [08:45<05:10, 57.33it/s]




 64%|██████▍   | 32205/50000 [08:45<05:34, 53.14it/s]




 64%|██████▍   | 32211/50000 [08:45<05:45, 51.46it/s]




 64%|██████▍   | 32217/50000 [08:46<06:05, 48.61it/s]




 64%|██████▍   | 32222/50000 [08:46<06:07, 48.42it/s]




 64%|██████▍   | 32227/50000 [08:46<06:16, 47.19it/s]




 64%|██████▍   | 32232/50000 [08:46<06:16, 47.22it/s]




 64%|██████▍   | 32237/50000 [08:46<06:38, 44.60it/s]




 64%|██████▍   | 32242/50000 [08:46<06:30, 45.50it/s]




 64%|██████▍   | 32249/50000 [08:46<05:58, 49.57it/s]




 65%|██████▍   | 32255/50000 [08:46<06:04, 48.65it/s]




 65%|██████▍   | 32261/50000 [08:46<05:48, 50.83it/s]




 65%|██████▍   | 32267/50000 [08:47<05:51, 50.49it/s]




 65%|██████▍   | 32273/50000 [08:47<05:50, 50.64it/s]




 65%|██████▍   | 32279/50000 [08:47<05:36, 52.60it/s]




 65%|██████▍   | 32286/50000 [08:47<05:17, 55.79it/s]




 65%|██████▍   | 32293/50000 [08:47<05:00, 58.85it/s]




 65%|██████▍   | 32299/50000 [08:47<05:11, 56.80it/s]




 65%|██████▍   | 32305/50000 [08:47<05:07, 57.52it/s]




 65%|██████▍   | 32312/50000 [08:47<04:58, 59.17it/s]




 65%|██████▍   | 32318/50000 [08:47<05:11, 56.69it/s]




 65%|██████▍   | 32325/50000 [08:48<05:00, 58.80it/s]




 65%|██████▍   | 32331/50000 [08:48<05:09, 57.17it/s]




 65%|██████▍   | 32337/50000 [08:48<05:13, 56.40it/s]




 65%|██████▍   | 32344/50000 [08:48<04:57, 59.40it/s]




 65%|██████▍   | 32351/50000 [08:48<05:11, 56.69it/s]




 65%|██████▍   | 32357/50000 [08:48<05:54, 49.81it/s]




 65%|██████▍   | 32363/50000 [08:48<06:13, 47.20it/s]




 65%|██████▍   | 32368/50000 [08:48<06:31, 45.06it/s]




 65%|██████▍   | 32374/50000 [08:49<06:23, 45.97it/s]




 65%|██████▍   | 32379/50000 [08:49<06:35, 44.51it/s]




 65%|██████▍   | 32386/50000 [08:49<05:57, 49.26it/s]




 65%|██████▍   | 32394/50000 [08:49<05:22, 54.60it/s]




 65%|██████▍   | 32401/50000 [08:49<05:02, 58.13it/s]




 65%|██████▍   | 32408/50000 [08:49<04:58, 58.90it/s]




 65%|██████▍   | 32415/50000 [08:49<04:55, 59.59it/s]




 65%|██████▍   | 32422/50000 [08:49<04:43, 62.07it/s]




 65%|██████▍   | 32429/50000 [08:49<04:47, 61.02it/s]




 65%|██████▍   | 32437/50000 [08:50<04:36, 63.56it/s]




 65%|██████▍   | 32444/50000 [08:50<04:48, 60.82it/s]




 65%|██████▍   | 32451/50000 [08:50<04:54, 59.49it/s]




 65%|██████▍   | 32458/50000 [08:50<04:48, 60.88it/s]




 65%|██████▍   | 32466/50000 [08:50<04:33, 64.21it/s]




 65%|██████▍   | 32473/50000 [08:50<04:30, 64.80it/s]




 65%|██████▍   | 32480/50000 [08:50<04:37, 63.19it/s]




 65%|██████▍   | 32487/50000 [08:50<04:31, 64.42it/s]




 65%|██████▍   | 32494/50000 [08:50<04:46, 61.05it/s]




 65%|██████▌   | 32501/50000 [08:51<04:45, 61.36it/s]




 65%|██████▌   | 32508/50000 [08:51<04:40, 62.41it/s]




 65%|██████▌   | 32515/50000 [08:51<04:31, 64.46it/s]




 65%|██████▌   | 32522/50000 [08:51<04:36, 63.31it/s]




 65%|██████▌   | 32529/50000 [08:51<04:35, 63.37it/s]




 65%|██████▌   | 32536/50000 [08:51<04:28, 65.09it/s]




 65%|██████▌   | 32543/50000 [08:51<04:26, 65.42it/s]




 65%|██████▌   | 32550/50000 [08:51<04:22, 66.40it/s]




 65%|██████▌   | 32557/50000 [08:51<04:34, 63.64it/s]




 65%|██████▌   | 32564/50000 [08:52<04:29, 64.75it/s]




 65%|██████▌   | 32571/50000 [08:52<04:34, 63.59it/s]




 65%|██████▌   | 32579/50000 [08:52<04:22, 66.29it/s]




 65%|██████▌   | 32587/50000 [08:52<04:15, 68.23it/s]




 65%|██████▌   | 32594/50000 [08:52<04:29, 64.54it/s]




 65%|██████▌   | 32601/50000 [08:52<04:24, 65.76it/s]




 65%|██████▌   | 32608/50000 [08:52<04:32, 63.74it/s]




 65%|██████▌   | 32615/50000 [08:52<04:31, 64.11it/s]




 65%|██████▌   | 32623/50000 [08:52<04:26, 65.28it/s]




 65%|██████▌   | 32630/50000 [08:53<04:29, 64.48it/s]




 65%|██████▌   | 32637/50000 [08:53<04:30, 64.28it/s]




 65%|██████▌   | 32644/50000 [08:53<04:32, 63.61it/s]




 65%|██████▌   | 32651/50000 [08:53<04:39, 61.98it/s]




 65%|██████▌   | 32658/50000 [08:53<04:30, 64.06it/s]




 65%|██████▌   | 32666/50000 [08:53<04:20, 66.65it/s]




 65%|██████▌   | 32673/50000 [08:53<04:33, 63.45it/s]




 65%|██████▌   | 32680/50000 [08:53<04:29, 64.26it/s]




 65%|██████▌   | 32687/50000 [08:53<04:36, 62.57it/s]




 65%|██████▌   | 32694/50000 [08:54<04:28, 64.40it/s]




 65%|██████▌   | 32701/50000 [08:54<04:26, 64.94it/s]




 65%|██████▌   | 32708/50000 [08:54<04:24, 65.50it/s]




 65%|██████▌   | 32715/50000 [08:54<04:37, 62.38it/s]




 65%|██████▌   | 32722/50000 [08:54<04:33, 63.14it/s]




 65%|██████▌   | 32729/50000 [08:54<04:33, 63.17it/s]




 65%|██████▌   | 32736/50000 [08:54<04:40, 61.52it/s]




 65%|██████▌   | 32743/50000 [08:54<04:47, 60.11it/s]




 66%|██████▌   | 32750/50000 [08:54<04:51, 59.17it/s]




 66%|██████▌   | 32756/50000 [08:55<04:51, 59.11it/s]




 66%|██████▌   | 32763/50000 [08:55<04:48, 59.67it/s]




 66%|██████▌   | 32770/50000 [08:55<04:45, 60.39it/s]




 66%|██████▌   | 32777/50000 [08:55<04:45, 60.42it/s]




 66%|██████▌   | 32784/50000 [08:55<04:37, 62.05it/s]




 66%|██████▌   | 32791/50000 [08:55<04:28, 64.11it/s]




 66%|██████▌   | 32798/50000 [08:55<04:23, 65.27it/s]




 66%|██████▌   | 32805/50000 [08:55<04:30, 63.63it/s]




 66%|██████▌   | 32813/50000 [08:55<04:17, 66.82it/s]




 66%|██████▌   | 32820/50000 [08:56<04:42, 60.83it/s]




 66%|██████▌   | 32828/50000 [08:56<04:30, 63.56it/s]




 66%|██████▌   | 32835/50000 [08:56<04:40, 61.14it/s]




 66%|██████▌   | 32843/50000 [08:56<04:27, 64.25it/s]




 66%|██████▌   | 32851/50000 [08:56<04:19, 65.97it/s]




 66%|██████▌   | 32858/50000 [08:56<04:22, 65.31it/s]




 66%|██████▌   | 32865/50000 [08:56<04:33, 62.59it/s]




 66%|██████▌   | 32872/50000 [08:56<04:30, 63.29it/s]




 66%|██████▌   | 32879/50000 [08:56<04:47, 59.57it/s]




 66%|██████▌   | 32886/50000 [08:57<04:45, 59.85it/s]




 66%|██████▌   | 32893/50000 [08:57<04:48, 59.39it/s]




 66%|██████▌   | 32900/50000 [08:57<04:43, 60.34it/s]




 66%|██████▌   | 32907/50000 [08:57<04:42, 60.55it/s]




 66%|██████▌   | 32914/50000 [08:57<04:34, 62.31it/s]




 66%|██████▌   | 32921/50000 [08:57<04:30, 63.09it/s]




 66%|██████▌   | 32929/50000 [08:57<04:17, 66.24it/s]




 66%|██████▌   | 32936/50000 [08:57<04:20, 65.49it/s]




 66%|██████▌   | 32944/50000 [08:57<04:10, 68.07it/s]




 66%|██████▌   | 32951/50000 [08:58<04:21, 65.24it/s]




 66%|██████▌   | 32958/50000 [08:58<04:20, 65.44it/s]




 66%|██████▌   | 32966/50000 [08:58<04:11, 67.69it/s]




 66%|██████▌   | 32973/50000 [08:58<04:23, 64.71it/s]




 66%|██████▌   | 32980/50000 [08:58<04:17, 66.07it/s]




 66%|██████▌   | 32987/50000 [08:58<04:15, 66.63it/s]




 66%|██████▌   | 32994/50000 [08:58<04:25, 64.09it/s]




 66%|██████▌   | 33001/50000 [08:58<04:37, 61.24it/s]




 66%|██████▌   | 33008/50000 [08:59<04:33, 62.23it/s]




 66%|██████▌   | 33015/50000 [08:59<04:27, 63.55it/s]




 66%|██████▌   | 33022/50000 [08:59<04:27, 63.45it/s]




 66%|██████▌   | 33030/50000 [08:59<04:13, 66.85it/s]




 66%|██████▌   | 33037/50000 [08:59<04:25, 63.93it/s]




 66%|██████▌   | 33044/50000 [08:59<04:25, 63.89it/s]




 66%|██████▌   | 33051/50000 [08:59<04:29, 63.01it/s]




 66%|██████▌   | 33058/50000 [08:59<04:38, 60.78it/s]




 66%|██████▌   | 33065/50000 [08:59<04:36, 61.33it/s]




 66%|██████▌   | 33072/50000 [09:00<04:34, 61.73it/s]




 66%|██████▌   | 33079/50000 [09:00<04:30, 62.50it/s]




 66%|██████▌   | 33086/50000 [09:00<04:32, 62.06it/s]




 66%|██████▌   | 33093/50000 [09:00<04:31, 62.24it/s]




 66%|██████▌   | 33100/50000 [09:00<04:29, 62.70it/s]




 66%|██████▌   | 33107/50000 [09:00<04:30, 62.52it/s]




 66%|██████▌   | 33114/50000 [09:00<04:34, 61.58it/s]




 66%|██████▌   | 33121/50000 [09:00<04:37, 60.93it/s]




 66%|██████▋   | 33128/50000 [09:00<04:38, 60.65it/s]




 66%|██████▋   | 33135/50000 [09:01<04:39, 60.45it/s]




 66%|██████▋   | 33142/50000 [09:01<04:33, 61.74it/s]




 66%|██████▋   | 33149/50000 [09:01<04:26, 63.19it/s]




 66%|██████▋   | 33156/50000 [09:01<04:23, 63.81it/s]




 66%|██████▋   | 33163/50000 [09:01<04:38, 60.42it/s]




 66%|██████▋   | 33170/50000 [09:01<05:15, 53.41it/s]




 66%|██████▋   | 33176/50000 [09:01<05:22, 52.25it/s]




 66%|██████▋   | 33182/50000 [09:01<05:40, 49.37it/s]




 66%|██████▋   | 33188/50000 [09:02<06:14, 44.93it/s]




 66%|██████▋   | 33193/50000 [09:02<06:08, 45.61it/s]




 66%|██████▋   | 33198/50000 [09:02<06:08, 45.60it/s]




 66%|██████▋   | 33204/50000 [09:02<05:49, 48.00it/s]




 66%|██████▋   | 33209/50000 [09:02<05:54, 47.38it/s]




 66%|██████▋   | 33214/50000 [09:02<06:05, 45.92it/s]




 66%|██████▋   | 33219/50000 [09:02<06:07, 45.69it/s]




 66%|██████▋   | 33225/50000 [09:02<05:41, 49.06it/s]




 66%|██████▋   | 33231/50000 [09:02<05:54, 47.34it/s]




 66%|██████▋   | 33238/50000 [09:03<05:21, 52.11it/s]




 66%|██████▋   | 33244/50000 [09:03<05:13, 53.42it/s]




 67%|██████▋   | 33251/50000 [09:03<04:57, 56.39it/s]




 67%|██████▋   | 33258/50000 [09:03<04:41, 59.39it/s]




 67%|██████▋   | 33265/50000 [09:03<04:36, 60.57it/s]




 67%|██████▋   | 33272/50000 [09:03<04:40, 59.55it/s]




 67%|██████▋   | 33279/50000 [09:03<04:49, 57.76it/s]




 67%|██████▋   | 33287/50000 [09:03<04:30, 61.89it/s]




 67%|██████▋   | 33294/50000 [09:03<04:23, 63.47it/s]




 67%|██████▋   | 33301/50000 [09:04<04:34, 60.76it/s]




 67%|██████▋   | 33308/50000 [09:04<04:51, 57.33it/s]




 67%|██████▋   | 33314/50000 [09:04<04:52, 57.10it/s]




 67%|██████▋   | 33320/50000 [09:04<04:52, 56.97it/s]




 67%|██████▋   | 33326/50000 [09:04<05:04, 54.81it/s]




 67%|██████▋   | 33332/50000 [09:04<05:05, 54.47it/s]




 67%|██████▋   | 33338/50000 [09:04<05:12, 53.38it/s]




 67%|██████▋   | 33344/50000 [09:04<05:43, 48.44it/s]




 67%|██████▋   | 33350/50000 [09:05<05:36, 49.48it/s]




 67%|██████▋   | 33356/50000 [09:05<05:31, 50.23it/s]




 67%|██████▋   | 33364/50000 [09:05<04:54, 56.39it/s]




 67%|██████▋   | 33370/50000 [09:05<04:59, 55.48it/s]




 67%|██████▋   | 33376/50000 [09:05<04:54, 56.49it/s]




 67%|██████▋   | 33382/50000 [09:05<04:53, 56.57it/s]




 67%|██████▋   | 33388/50000 [09:05<04:58, 55.68it/s]




 67%|██████▋   | 33395/50000 [09:05<04:49, 57.33it/s]




 67%|██████▋   | 33402/50000 [09:05<04:41, 58.98it/s]




 67%|██████▋   | 33409/50000 [09:06<04:37, 59.73it/s]




 67%|██████▋   | 33417/50000 [09:06<04:23, 63.01it/s]




 67%|██████▋   | 33425/50000 [09:06<04:11, 65.85it/s]




 67%|██████▋   | 33432/50000 [09:06<04:13, 65.28it/s]




 67%|██████▋   | 33439/50000 [09:06<04:21, 63.42it/s]




 67%|██████▋   | 33446/50000 [09:06<04:21, 63.24it/s]




 67%|██████▋   | 33453/50000 [09:06<04:19, 63.66it/s]




 67%|██████▋   | 33460/50000 [09:06<04:19, 63.79it/s]




 67%|██████▋   | 33467/50000 [09:06<04:15, 64.68it/s]




 67%|██████▋   | 33474/50000 [09:07<04:29, 61.21it/s]




 67%|██████▋   | 33481/50000 [09:07<04:23, 62.80it/s]




 67%|██████▋   | 33488/50000 [09:07<04:16, 64.31it/s]




 67%|██████▋   | 33496/50000 [09:07<04:06, 66.84it/s]




 67%|██████▋   | 33503/50000 [09:07<04:15, 64.45it/s]




 67%|██████▋   | 33511/50000 [09:07<04:03, 67.62it/s]




 67%|██████▋   | 33518/50000 [09:07<04:14, 64.77it/s]




 67%|██████▋   | 33526/50000 [09:07<04:08, 66.19it/s]




 67%|██████▋   | 33533/50000 [09:07<04:08, 66.17it/s]




 67%|██████▋   | 33540/50000 [09:08<04:10, 65.81it/s]




 67%|██████▋   | 33547/50000 [09:08<04:18, 63.60it/s]




 67%|██████▋   | 33554/50000 [09:08<04:23, 62.30it/s]




 67%|██████▋   | 33562/50000 [09:08<04:09, 65.87it/s]




 67%|██████▋   | 33569/50000 [09:08<04:19, 63.29it/s]




 67%|██████▋   | 33576/50000 [09:08<04:20, 63.01it/s]




 67%|██████▋   | 33583/50000 [09:08<04:21, 62.72it/s]




 67%|██████▋   | 33590/50000 [09:08<04:13, 64.61it/s]




 67%|██████▋   | 33597/50000 [09:08<04:19, 63.15it/s]




 67%|██████▋   | 33604/50000 [09:09<04:18, 63.35it/s]




 67%|██████▋   | 33611/50000 [09:09<04:23, 62.13it/s]




 67%|██████▋   | 33618/50000 [09:09<04:25, 61.64it/s]




 67%|██████▋   | 33625/50000 [09:09<04:29, 60.82it/s]




 67%|██████▋   | 33632/50000 [09:09<04:32, 60.10it/s]




 67%|██████▋   | 33640/50000 [09:09<04:16, 63.67it/s]




 67%|██████▋   | 33648/50000 [09:09<04:05, 66.60it/s]




 67%|██████▋   | 33655/50000 [09:09<04:03, 67.06it/s]




 67%|██████▋   | 33662/50000 [09:09<04:06, 66.23it/s]




 67%|██████▋   | 33669/50000 [09:10<04:22, 62.18it/s]




 67%|██████▋   | 33676/50000 [09:10<04:24, 61.83it/s]




 67%|██████▋   | 33684/50000 [09:10<04:18, 63.10it/s]




 67%|██████▋   | 33693/50000 [09:10<03:58, 68.49it/s]




 67%|██████▋   | 33701/50000 [09:10<04:00, 67.90it/s]




 67%|██████▋   | 33708/50000 [09:10<04:12, 64.59it/s]




 67%|██████▋   | 33715/50000 [09:10<04:22, 62.13it/s]




 67%|██████▋   | 33722/50000 [09:10<04:23, 61.80it/s]




 67%|██████▋   | 33729/50000 [09:11<04:22, 62.06it/s]




 67%|██████▋   | 33736/50000 [09:11<04:29, 60.31it/s]




 67%|██████▋   | 33743/50000 [09:11<04:34, 59.30it/s]




 67%|██████▋   | 33749/50000 [09:11<04:48, 56.37it/s]




 68%|██████▊   | 33755/50000 [09:11<04:48, 56.30it/s]




 68%|██████▊   | 33762/50000 [09:11<04:37, 58.57it/s]




 68%|██████▊   | 33768/50000 [09:11<04:47, 56.54it/s]




 68%|██████▊   | 33774/50000 [09:11<05:02, 53.58it/s]




 68%|██████▊   | 33781/50000 [09:11<04:51, 55.62it/s]




 68%|██████▊   | 33788/50000 [09:12<04:36, 58.72it/s]




 68%|██████▊   | 33795/50000 [09:12<04:22, 61.66it/s]




 68%|██████▊   | 33802/50000 [09:12<04:20, 62.20it/s]




 68%|██████▊   | 33809/50000 [09:12<04:21, 62.01it/s]




 68%|██████▊   | 33816/50000 [09:12<04:18, 62.54it/s]




 68%|██████▊   | 33823/50000 [09:12<04:12, 63.95it/s]




 68%|██████▊   | 33830/50000 [09:12<04:15, 63.38it/s]




 68%|██████▊   | 33837/50000 [09:12<04:15, 63.17it/s]




 68%|██████▊   | 33844/50000 [09:12<04:12, 64.05it/s]




 68%|██████▊   | 33851/50000 [09:13<04:20, 61.94it/s]




 68%|██████▊   | 33858/50000 [09:13<04:14, 63.34it/s]




 68%|██████▊   | 33865/50000 [09:13<04:16, 62.80it/s]




 68%|██████▊   | 33872/50000 [09:13<04:13, 63.61it/s]




 68%|██████▊   | 33879/50000 [09:13<04:16, 62.81it/s]




 68%|██████▊   | 33886/50000 [09:13<04:36, 58.38it/s]




 68%|██████▊   | 33893/50000 [09:13<04:25, 60.77it/s]




 68%|██████▊   | 33900/50000 [09:13<04:18, 62.38it/s]




 68%|██████▊   | 33907/50000 [09:13<04:14, 63.31it/s]




 68%|██████▊   | 33915/50000 [09:14<04:03, 66.08it/s]




 68%|██████▊   | 33922/50000 [09:14<04:09, 64.48it/s]




 68%|██████▊   | 33929/50000 [09:14<04:12, 63.58it/s]




 68%|██████▊   | 33936/50000 [09:14<04:17, 62.29it/s]




 68%|██████▊   | 33943/50000 [09:14<04:11, 63.76it/s]




 68%|██████▊   | 33951/50000 [09:14<03:59, 67.09it/s]




 68%|██████▊   | 33958/50000 [09:14<04:06, 65.15it/s]




 68%|██████▊   | 33965/50000 [09:14<04:11, 63.86it/s]




 68%|██████▊   | 33972/50000 [09:14<04:17, 62.15it/s]




 68%|██████▊   | 33979/50000 [09:15<04:17, 62.22it/s]




 68%|██████▊   | 33986/50000 [09:15<04:14, 62.94it/s]




 68%|██████▊   | 33993/50000 [09:15<04:12, 63.37it/s]




 68%|██████▊   | 34000/50000 [09:15<04:22, 61.01it/s]




 68%|██████▊   | 34007/50000 [09:15<04:19, 61.66it/s]




 68%|██████▊   | 34014/50000 [09:15<04:17, 62.13it/s]




 68%|██████▊   | 34022/50000 [09:15<04:00, 66.30it/s]




 68%|██████▊   | 34029/50000 [09:15<04:03, 65.72it/s]




 68%|██████▊   | 34037/50000 [09:15<03:53, 68.24it/s]




 68%|██████▊   | 34044/50000 [09:16<04:07, 64.45it/s]




 68%|██████▊   | 34051/50000 [09:16<04:09, 63.91it/s]




 68%|██████▊   | 34058/50000 [09:16<04:17, 61.85it/s]




 68%|██████▊   | 34066/50000 [09:16<04:01, 66.08it/s]




 68%|██████▊   | 34073/50000 [09:16<04:06, 64.55it/s]




 68%|██████▊   | 34080/50000 [09:16<04:08, 64.15it/s]




 68%|██████▊   | 34088/50000 [09:16<03:57, 67.13it/s]




 68%|██████▊   | 34095/50000 [09:16<04:17, 61.73it/s]




 68%|██████▊   | 34103/50000 [09:16<04:01, 65.83it/s]




 68%|██████▊   | 34110/50000 [09:17<04:02, 65.57it/s]




 68%|██████▊   | 34117/50000 [09:17<03:58, 66.70it/s]




 68%|██████▊   | 34124/50000 [09:17<04:00, 65.99it/s]




 68%|██████▊   | 34131/50000 [09:17<04:00, 65.87it/s]




 68%|██████▊   | 34138/50000 [09:17<04:32, 58.25it/s]




 68%|██████▊   | 34145/50000 [09:17<04:33, 57.89it/s]




 68%|██████▊   | 34151/50000 [09:17<04:53, 53.91it/s]




 68%|██████▊   | 34157/50000 [09:17<05:02, 52.44it/s]




 68%|██████▊   | 34163/50000 [09:18<05:20, 49.43it/s]




 68%|██████▊   | 34169/50000 [09:18<05:15, 50.14it/s]




 68%|██████▊   | 34175/50000 [09:18<05:09, 51.09it/s]




 68%|██████▊   | 34181/50000 [09:18<05:02, 52.32it/s]




 68%|██████▊   | 34187/50000 [09:18<05:12, 50.54it/s]




 68%|██████▊   | 34193/50000 [09:18<05:02, 52.32it/s]




 68%|██████▊   | 34199/50000 [09:18<05:18, 49.66it/s]




 68%|██████▊   | 34205/50000 [09:18<05:11, 50.68it/s]




 68%|██████▊   | 34211/50000 [09:18<05:02, 52.23it/s]




 68%|██████▊   | 34218/50000 [09:19<04:43, 55.76it/s]




 68%|██████▊   | 34224/50000 [09:19<04:47, 54.83it/s]




 68%|██████▊   | 34232/50000 [09:19<04:27, 58.85it/s]




 68%|██████▊   | 34239/50000 [09:19<04:26, 59.19it/s]




 68%|██████▊   | 34247/50000 [09:19<04:15, 61.72it/s]




 69%|██████▊   | 34254/50000 [09:19<04:23, 59.68it/s]




 69%|██████▊   | 34261/50000 [09:19<04:13, 61.99it/s]




 69%|██████▊   | 34269/50000 [09:19<03:57, 66.27it/s]




 69%|██████▊   | 34276/50000 [09:20<04:08, 63.38it/s]




 69%|██████▊   | 34283/50000 [09:20<04:22, 59.91it/s]




 69%|██████▊   | 34290/50000 [09:20<04:15, 61.51it/s]




 69%|██████▊   | 34297/50000 [09:20<04:35, 57.02it/s]




 69%|██████▊   | 34303/50000 [09:20<04:55, 53.17it/s]




 69%|██████▊   | 34309/50000 [09:20<05:31, 47.40it/s]




 69%|██████▊   | 34314/50000 [09:20<06:08, 42.55it/s]




 69%|██████▊   | 34320/50000 [09:20<05:42, 45.79it/s]




 69%|██████▊   | 34326/50000 [09:21<05:21, 48.73it/s]




 69%|██████▊   | 34332/50000 [09:21<05:07, 51.02it/s]




 69%|██████▊   | 34338/50000 [09:21<04:56, 52.75it/s]




 69%|██████▊   | 34344/50000 [09:21<05:07, 50.95it/s]




 69%|██████▊   | 34351/50000 [09:21<04:45, 54.73it/s]




 69%|██████▊   | 34357/50000 [09:21<04:58, 52.46it/s]




 69%|██████▊   | 34363/50000 [09:21<04:52, 53.54it/s]




 69%|██████▊   | 34370/50000 [09:21<04:36, 56.59it/s]




 69%|██████▉   | 34376/50000 [09:21<04:47, 54.26it/s]




 69%|██████▉   | 34382/50000 [09:22<04:43, 55.06it/s]




 69%|██████▉   | 34389/50000 [09:22<04:27, 58.35it/s]




 69%|██████▉   | 34397/50000 [09:22<04:07, 63.02it/s]




 69%|██████▉   | 34404/50000 [09:22<04:01, 64.47it/s]




 69%|██████▉   | 34411/50000 [09:22<04:09, 62.39it/s]




 69%|██████▉   | 34418/50000 [09:22<04:09, 62.47it/s]




 69%|██████▉   | 34425/50000 [09:22<04:03, 63.90it/s]




 69%|██████▉   | 34432/50000 [09:22<04:19, 59.94it/s]




 69%|██████▉   | 34439/50000 [09:22<04:30, 57.45it/s]




 69%|██████▉   | 34445/50000 [09:23<04:37, 56.12it/s]




 69%|██████▉   | 34452/50000 [09:23<04:26, 58.29it/s]




 69%|██████▉   | 34458/50000 [09:23<04:32, 57.00it/s]




 69%|██████▉   | 34464/50000 [09:23<04:31, 57.25it/s]




 69%|██████▉   | 34472/50000 [09:23<04:12, 61.43it/s]




 69%|██████▉   | 34480/50000 [09:23<04:00, 64.48it/s]




 69%|██████▉   | 34487/50000 [09:23<04:02, 64.10it/s]




 69%|██████▉   | 34494/50000 [09:23<04:12, 61.49it/s]




 69%|██████▉   | 34501/50000 [09:23<04:17, 60.25it/s]




 69%|██████▉   | 34508/50000 [09:24<04:08, 62.25it/s]




 69%|██████▉   | 34516/50000 [09:24<03:55, 65.75it/s]




 69%|██████▉   | 34524/50000 [09:24<03:48, 67.75it/s]




 69%|██████▉   | 34533/50000 [09:24<03:35, 71.82it/s]




 69%|██████▉   | 34541/50000 [09:24<03:40, 70.07it/s]




 69%|██████▉   | 34549/50000 [09:24<03:55, 65.68it/s]




 69%|██████▉   | 34556/50000 [09:24<03:58, 64.74it/s]




 69%|██████▉   | 34563/50000 [09:24<03:58, 64.79it/s]




 69%|██████▉   | 34570/50000 [09:24<04:03, 63.35it/s]




 69%|██████▉   | 34577/50000 [09:25<04:07, 62.22it/s]




 69%|██████▉   | 34584/50000 [09:25<04:06, 62.52it/s]




 69%|██████▉   | 34591/50000 [09:25<04:05, 62.73it/s]




 69%|██████▉   | 34598/50000 [09:25<04:05, 62.71it/s]




 69%|██████▉   | 34606/50000 [09:25<03:54, 65.62it/s]




 69%|██████▉   | 34613/50000 [09:25<03:53, 65.98it/s]




 69%|██████▉   | 34620/50000 [09:25<03:56, 65.13it/s]




 69%|██████▉   | 34627/50000 [09:25<04:02, 63.33it/s]




 69%|██████▉   | 34634/50000 [09:25<04:00, 63.99it/s]




 69%|██████▉   | 34642/50000 [09:26<03:56, 64.98it/s]




 69%|██████▉   | 34649/50000 [09:26<03:53, 65.75it/s]




 69%|██████▉   | 34656/50000 [09:26<03:50, 66.45it/s]




 69%|██████▉   | 34663/50000 [09:26<03:49, 66.76it/s]




 69%|██████▉   | 34671/50000 [09:26<03:42, 68.76it/s]




 69%|██████▉   | 34679/50000 [09:26<03:40, 69.59it/s]




 69%|██████▉   | 34686/50000 [09:26<03:46, 67.75it/s]




 69%|██████▉   | 34693/50000 [09:26<03:44, 68.26it/s]




 69%|██████▉   | 34701/50000 [09:26<03:40, 69.42it/s]




 69%|██████▉   | 34709/50000 [09:27<03:36, 70.62it/s]




 69%|██████▉   | 34717/50000 [09:27<03:43, 68.38it/s]




 69%|██████▉   | 34724/50000 [09:27<03:46, 67.52it/s]




 69%|██████▉   | 34731/50000 [09:27<03:52, 65.80it/s]




 69%|██████▉   | 34739/50000 [09:27<03:43, 68.13it/s]




 69%|██████▉   | 34746/50000 [09:27<03:57, 64.21it/s]




 70%|██████▉   | 34753/50000 [09:27<04:08, 61.24it/s]




 70%|██████▉   | 34760/50000 [09:27<04:05, 62.20it/s]




 70%|██████▉   | 34769/50000 [09:27<03:46, 67.36it/s]




 70%|██████▉   | 34776/50000 [09:28<03:51, 65.88it/s]




 70%|██████▉   | 34783/50000 [09:28<03:55, 64.70it/s]




 70%|██████▉   | 34790/50000 [09:28<04:05, 62.04it/s]




 70%|██████▉   | 34797/50000 [09:28<04:12, 60.30it/s]




 70%|██████▉   | 34804/50000 [09:28<04:07, 61.31it/s]




 70%|██████▉   | 34812/50000 [09:28<03:54, 64.70it/s]




 70%|██████▉   | 34819/50000 [09:28<03:49, 66.07it/s]




 70%|██████▉   | 34826/50000 [09:28<03:46, 66.87it/s]




 70%|██████▉   | 34833/50000 [09:28<03:50, 65.92it/s]




 70%|██████▉   | 34840/50000 [09:29<03:54, 64.55it/s]




 70%|██████▉   | 34847/50000 [09:29<04:02, 62.60it/s]




 70%|██████▉   | 34854/50000 [09:29<03:55, 64.29it/s]




 70%|██████▉   | 34861/50000 [09:29<03:52, 65.16it/s]




 70%|██████▉   | 34868/50000 [09:29<03:52, 65.02it/s]




 70%|██████▉   | 34875/50000 [09:29<03:52, 65.10it/s]




 70%|██████▉   | 34883/50000 [09:29<03:45, 66.93it/s]




 70%|██████▉   | 34890/50000 [09:29<03:43, 67.48it/s]




 70%|██████▉   | 34897/50000 [09:29<03:43, 67.68it/s]




 70%|██████▉   | 34904/50000 [09:30<03:43, 67.62it/s]




 70%|██████▉   | 34912/50000 [09:30<03:36, 69.85it/s]




 70%|██████▉   | 34920/50000 [09:30<03:37, 69.28it/s]




 70%|██████▉   | 34927/50000 [09:30<03:41, 68.13it/s]




 70%|██████▉   | 34934/50000 [09:30<03:58, 63.08it/s]




 70%|██████▉   | 34941/50000 [09:30<03:52, 64.70it/s]




 70%|██████▉   | 34948/50000 [09:30<03:56, 63.77it/s]




 70%|██████▉   | 34955/50000 [09:30<04:04, 61.50it/s]




 70%|██████▉   | 34962/50000 [09:30<04:05, 61.24it/s]




 70%|██████▉   | 34969/50000 [09:31<04:04, 61.50it/s]




 70%|██████▉   | 34977/50000 [09:31<03:51, 65.00it/s]




 70%|██████▉   | 34984/50000 [09:31<04:04, 61.41it/s]




 70%|██████▉   | 34991/50000 [09:31<04:16, 58.54it/s]




 70%|██████▉   | 34998/50000 [09:31<04:13, 59.26it/s]




 70%|███████   | 35005/50000 [09:31<04:03, 61.68it/s]




 70%|███████   | 35012/50000 [09:31<04:03, 61.48it/s]




 70%|███████   | 35019/50000 [09:31<04:03, 61.51it/s]




 70%|███████   | 35026/50000 [09:32<04:08, 60.18it/s]




 70%|███████   | 35033/50000 [09:32<04:08, 60.20it/s]




 70%|███████   | 35040/50000 [09:32<04:04, 61.08it/s]




 70%|███████   | 35047/50000 [09:32<03:59, 62.37it/s]




 70%|███████   | 35054/50000 [09:32<04:01, 61.80it/s]




 70%|███████   | 35061/50000 [09:32<03:56, 63.14it/s]




 70%|███████   | 35068/50000 [09:32<03:57, 62.97it/s]




 70%|███████   | 35075/50000 [09:32<03:51, 64.35it/s]




 70%|███████   | 35082/50000 [09:32<03:55, 63.43it/s]




 70%|███████   | 35089/50000 [09:33<03:59, 62.27it/s]




 70%|███████   | 35097/50000 [09:33<03:44, 66.42it/s]




 70%|███████   | 35104/50000 [09:33<03:48, 65.06it/s]




 70%|███████   | 35111/50000 [09:33<03:45, 65.96it/s]




 70%|███████   | 35118/50000 [09:33<03:44, 66.41it/s]




 70%|███████   | 35125/50000 [09:33<04:05, 60.52it/s]




 70%|███████   | 35132/50000 [09:33<04:25, 56.02it/s]




 70%|███████   | 35138/50000 [09:33<04:39, 53.11it/s]




 70%|███████   | 35144/50000 [09:33<04:47, 51.64it/s]




 70%|███████   | 35150/50000 [09:34<04:57, 49.91it/s]




 70%|███████   | 35156/50000 [09:34<05:01, 49.24it/s]




 70%|███████   | 35161/50000 [09:34<05:18, 46.60it/s]




 70%|███████   | 35167/50000 [09:34<05:06, 48.47it/s]




 70%|███████   | 35172/50000 [09:34<05:04, 48.64it/s]




 70%|███████   | 35177/50000 [09:34<05:09, 47.82it/s]




 70%|███████   | 35182/50000 [09:34<05:19, 46.40it/s]




 70%|███████   | 35187/50000 [09:34<05:20, 46.15it/s]




 70%|███████   | 35193/50000 [09:35<05:06, 48.30it/s]




 70%|███████   | 35198/50000 [09:35<05:04, 48.55it/s]




 70%|███████   | 35205/50000 [09:35<04:40, 52.66it/s]




 70%|███████   | 35211/50000 [09:35<04:32, 54.26it/s]




 70%|███████   | 35217/50000 [09:35<04:33, 54.10it/s]




 70%|███████   | 35223/50000 [09:35<04:25, 55.63it/s]




 70%|███████   | 35230/50000 [09:35<04:17, 57.43it/s]




 70%|███████   | 35238/50000 [09:35<04:05, 60.10it/s]




 70%|███████   | 35245/50000 [09:35<03:57, 62.14it/s]




 71%|███████   | 35252/50000 [09:35<04:03, 60.68it/s]




 71%|███████   | 35259/50000 [09:36<04:01, 60.94it/s]




 71%|███████   | 35266/50000 [09:36<04:34, 53.75it/s]




 71%|███████   | 35273/50000 [09:36<04:19, 56.80it/s]




 71%|███████   | 35279/50000 [09:36<04:45, 51.48it/s]




 71%|███████   | 35285/50000 [09:36<05:09, 47.56it/s]




 71%|███████   | 35290/50000 [09:36<05:23, 45.41it/s]




 71%|███████   | 35295/50000 [09:36<05:46, 42.46it/s]




 71%|███████   | 35300/50000 [09:37<05:41, 43.09it/s]




 71%|███████   | 35307/50000 [09:37<05:07, 47.76it/s]




 71%|███████   | 35313/50000 [09:37<05:10, 47.30it/s]




 71%|███████   | 35321/50000 [09:37<04:38, 52.77it/s]




 71%|███████   | 35329/50000 [09:37<04:19, 56.51it/s]




 71%|███████   | 35337/50000 [09:37<04:01, 60.66it/s]




 71%|███████   | 35344/50000 [09:37<04:05, 59.61it/s]




 71%|███████   | 35351/50000 [09:37<04:08, 58.97it/s]




 71%|███████   | 35358/50000 [09:37<04:03, 60.03it/s]




 71%|███████   | 35365/50000 [09:38<04:05, 59.71it/s]




 71%|███████   | 35373/50000 [09:38<03:50, 63.44it/s]




 71%|███████   | 35380/50000 [09:38<03:49, 63.73it/s]




 71%|███████   | 35387/50000 [09:38<03:55, 62.06it/s]




 71%|███████   | 35394/50000 [09:38<04:02, 60.16it/s]




 71%|███████   | 35401/50000 [09:38<04:07, 58.97it/s]




 71%|███████   | 35407/50000 [09:38<04:19, 56.33it/s]




 71%|███████   | 35414/50000 [09:38<04:06, 59.12it/s]




 71%|███████   | 35421/50000 [09:39<04:00, 60.61it/s]




 71%|███████   | 35428/50000 [09:39<03:56, 61.63it/s]




 71%|███████   | 35435/50000 [09:39<04:01, 60.30it/s]




 71%|███████   | 35442/50000 [09:39<04:06, 58.99it/s]




 71%|███████   | 35448/50000 [09:39<04:06, 58.99it/s]




 71%|███████   | 35454/50000 [09:39<04:05, 59.16it/s]




 71%|███████   | 35460/50000 [09:39<04:20, 55.82it/s]




 71%|███████   | 35466/50000 [09:39<04:15, 56.89it/s]




 71%|███████   | 35472/50000 [09:39<04:12, 57.50it/s]




 71%|███████   | 35478/50000 [09:39<04:09, 58.10it/s]




 71%|███████   | 35484/50000 [09:40<04:22, 55.31it/s]




 71%|███████   | 35491/50000 [09:40<04:15, 56.77it/s]




 71%|███████   | 35497/50000 [09:40<04:22, 55.27it/s]




 71%|███████   | 35503/50000 [09:40<04:18, 56.17it/s]




 71%|███████   | 35510/50000 [09:40<04:04, 59.15it/s]




 71%|███████   | 35518/50000 [09:40<03:50, 62.85it/s]




 71%|███████   | 35525/50000 [09:40<03:50, 62.87it/s]




 71%|███████   | 35532/50000 [09:40<03:56, 61.25it/s]




 71%|███████   | 35539/50000 [09:41<04:01, 59.93it/s]




 71%|███████   | 35546/50000 [09:41<04:08, 58.16it/s]




 71%|███████   | 35552/50000 [09:41<04:13, 56.91it/s]




 71%|███████   | 35559/50000 [09:41<04:03, 59.27it/s]




 71%|███████   | 35566/50000 [09:41<04:01, 59.78it/s]




 71%|███████   | 35573/50000 [09:41<03:54, 61.42it/s]




 71%|███████   | 35580/50000 [09:41<04:01, 59.74it/s]




 71%|███████   | 35587/50000 [09:41<03:59, 60.27it/s]




 71%|███████   | 35595/50000 [09:41<03:43, 64.50it/s]




 71%|███████   | 35602/50000 [09:42<03:42, 64.83it/s]




 71%|███████   | 35609/50000 [09:42<03:59, 60.05it/s]




 71%|███████   | 35616/50000 [09:42<04:15, 56.38it/s]




 71%|███████   | 35622/50000 [09:42<04:26, 53.91it/s]




 71%|███████▏  | 35629/50000 [09:42<04:14, 56.54it/s]




 71%|███████▏  | 35635/50000 [09:42<04:33, 52.46it/s]




 71%|███████▏  | 35642/50000 [09:42<04:19, 55.28it/s]




 71%|███████▏  | 35648/50000 [09:42<04:24, 54.36it/s]




 71%|███████▏  | 35655/50000 [09:43<04:09, 57.44it/s]




 71%|███████▏  | 35661/50000 [09:43<04:19, 55.19it/s]




 71%|███████▏  | 35668/50000 [09:43<04:11, 57.10it/s]




 71%|███████▏  | 35676/50000 [09:43<03:51, 61.92it/s]




 71%|███████▏  | 35683/50000 [09:43<03:59, 59.76it/s]




 71%|███████▏  | 35690/50000 [09:43<03:56, 60.55it/s]




 71%|███████▏  | 35697/50000 [09:43<04:05, 58.27it/s]




 71%|███████▏  | 35703/50000 [09:43<04:16, 55.72it/s]




 71%|███████▏  | 35709/50000 [09:43<04:15, 55.95it/s]




 71%|███████▏  | 35715/50000 [09:44<04:36, 51.65it/s]




 71%|███████▏  | 35722/50000 [09:44<04:16, 55.57it/s]




 71%|███████▏  | 35728/50000 [09:44<04:19, 55.00it/s]




 71%|███████▏  | 35734/50000 [09:44<04:29, 53.02it/s]




 71%|███████▏  | 35740/50000 [09:44<04:22, 54.23it/s]




 71%|███████▏  | 35746/50000 [09:44<04:29, 52.93it/s]




 72%|███████▏  | 35752/50000 [09:44<04:30, 52.75it/s]




 72%|███████▏  | 35758/50000 [09:44<04:24, 53.74it/s]




 72%|███████▏  | 35764/50000 [09:44<04:24, 53.74it/s]




 72%|███████▏  | 35770/50000 [09:45<04:17, 55.21it/s]




 72%|███████▏  | 35776/50000 [09:45<04:21, 54.31it/s]




 72%|███████▏  | 35782/50000 [09:45<04:21, 54.42it/s]




 72%|███████▏  | 35789/50000 [09:45<04:08, 57.21it/s]




 72%|███████▏  | 35795/50000 [09:45<04:17, 55.25it/s]




 72%|███████▏  | 35802/50000 [09:45<04:06, 57.53it/s]




 72%|███████▏  | 35808/50000 [09:45<04:10, 56.67it/s]




 72%|███████▏  | 35815/50000 [09:45<04:00, 59.08it/s]




 72%|███████▏  | 35822/50000 [09:45<03:49, 61.86it/s]




 72%|███████▏  | 35830/50000 [09:46<03:37, 65.28it/s]




 72%|███████▏  | 35838/50000 [09:46<03:29, 67.57it/s]




 72%|███████▏  | 35845/50000 [09:46<03:28, 67.94it/s]




 72%|███████▏  | 35852/50000 [09:46<03:44, 62.88it/s]




 72%|███████▏  | 35859/50000 [09:46<03:40, 64.20it/s]




 72%|███████▏  | 35866/50000 [09:46<03:35, 65.70it/s]




 72%|███████▏  | 35873/50000 [09:46<03:37, 64.94it/s]




 72%|███████▏  | 35881/50000 [09:46<03:30, 67.14it/s]




 72%|███████▏  | 35889/50000 [09:46<03:25, 68.78it/s]




 72%|███████▏  | 35897/50000 [09:47<03:24, 68.89it/s]




 72%|███████▏  | 35904/50000 [09:47<03:27, 67.94it/s]




 72%|███████▏  | 35911/50000 [09:47<03:28, 67.71it/s]




 72%|███████▏  | 35918/50000 [09:47<03:30, 66.87it/s]




 72%|███████▏  | 35925/50000 [09:47<03:47, 61.91it/s]




 72%|███████▏  | 35932/50000 [09:47<03:40, 63.83it/s]




 72%|███████▏  | 35939/50000 [09:47<03:43, 62.96it/s]




 72%|███████▏  | 35947/50000 [09:47<03:34, 65.65it/s]




 72%|███████▏  | 35954/50000 [09:47<03:38, 64.37it/s]




 72%|███████▏  | 35961/50000 [09:48<03:39, 63.85it/s]




 72%|███████▏  | 35968/50000 [09:48<03:41, 63.22it/s]




 72%|███████▏  | 35975/50000 [09:48<03:42, 62.97it/s]




 72%|███████▏  | 35983/50000 [09:48<03:38, 64.24it/s]




 72%|███████▏  | 35990/50000 [09:48<03:36, 64.82it/s]




 72%|███████▏  | 35997/50000 [09:48<03:35, 64.87it/s]




 72%|███████▏  | 36004/50000 [09:48<03:32, 65.82it/s]




 72%|███████▏  | 36011/50000 [09:48<03:32, 65.69it/s]




 72%|███████▏  | 36018/50000 [09:48<03:32, 65.75it/s]




 72%|███████▏  | 36026/50000 [09:49<03:23, 68.71it/s]




 72%|███████▏  | 36035/50000 [09:49<03:12, 72.49it/s]




 72%|███████▏  | 36043/50000 [09:49<03:35, 64.63it/s]




 72%|███████▏  | 36050/50000 [09:49<03:42, 62.82it/s]




 72%|███████▏  | 36057/50000 [09:49<04:01, 57.67it/s]




 72%|███████▏  | 36063/50000 [09:49<04:34, 50.72it/s]




 72%|███████▏  | 36069/50000 [09:49<04:28, 51.85it/s]




 72%|███████▏  | 36075/50000 [09:49<04:28, 51.85it/s]




 72%|███████▏  | 36081/50000 [09:50<04:47, 48.35it/s]




 72%|███████▏  | 36087/50000 [09:50<04:37, 50.12it/s]




 72%|███████▏  | 36093/50000 [09:50<04:35, 50.50it/s]




 72%|███████▏  | 36099/50000 [09:50<05:07, 45.17it/s]




 72%|███████▏  | 36104/50000 [09:50<05:00, 46.22it/s]




 72%|███████▏  | 36109/50000 [09:50<04:57, 46.67it/s]




 72%|███████▏  | 36114/50000 [09:50<05:10, 44.66it/s]




 72%|███████▏  | 36119/50000 [09:50<05:14, 44.16it/s]




 72%|███████▏  | 36124/50000 [09:51<05:11, 44.51it/s]




 72%|███████▏  | 36130/50000 [09:51<04:51, 47.59it/s]




 72%|███████▏  | 36138/50000 [09:51<04:18, 53.55it/s]




 72%|███████▏  | 36145/50000 [09:51<04:06, 56.13it/s]




 72%|███████▏  | 36152/50000 [09:51<03:52, 59.57it/s]




 72%|███████▏  | 36161/50000 [09:51<03:33, 64.90it/s]




 72%|███████▏  | 36168/50000 [09:51<03:31, 65.29it/s]




 72%|███████▏  | 36175/50000 [09:51<03:32, 65.02it/s]




 72%|███████▏  | 36182/50000 [09:51<03:34, 64.29it/s]




 72%|███████▏  | 36189/50000 [09:51<03:34, 64.32it/s]




 72%|███████▏  | 36196/50000 [09:52<03:45, 61.31it/s]




 72%|███████▏  | 36203/50000 [09:52<03:57, 58.20it/s]




 72%|███████▏  | 36211/50000 [09:52<03:42, 61.89it/s]




 72%|███████▏  | 36218/50000 [09:52<04:12, 54.52it/s]




 72%|███████▏  | 36224/50000 [09:52<04:25, 51.82it/s]




 72%|███████▏  | 36230/50000 [09:52<04:51, 47.20it/s]




 72%|███████▏  | 36235/50000 [09:52<05:01, 45.68it/s]




 72%|███████▏  | 36241/50000 [09:53<04:40, 48.99it/s]




 72%|███████▏  | 36247/50000 [09:53<04:35, 49.94it/s]




 73%|███████▎  | 36255/50000 [09:53<04:09, 55.04it/s]




 73%|███████▎  | 36262/50000 [09:53<03:58, 57.69it/s]




 73%|███████▎  | 36270/50000 [09:53<03:42, 61.69it/s]




 73%|███████▎  | 36278/50000 [09:53<03:30, 65.15it/s]




 73%|███████▎  | 36285/50000 [09:53<03:30, 65.10it/s]




 73%|███████▎  | 36292/50000 [09:53<03:26, 66.36it/s]




 73%|███████▎  | 36299/50000 [09:53<03:34, 63.79it/s]




 73%|███████▎  | 36306/50000 [09:54<03:39, 62.34it/s]




 73%|███████▎  | 36313/50000 [09:54<03:40, 62.11it/s]




 73%|███████▎  | 36320/50000 [09:54<03:36, 63.29it/s]




 73%|███████▎  | 36328/50000 [09:54<03:26, 66.06it/s]




 73%|███████▎  | 36335/50000 [09:54<03:27, 65.92it/s]




 73%|███████▎  | 36342/50000 [09:54<03:28, 65.45it/s]




 73%|███████▎  | 36349/50000 [09:54<03:35, 63.37it/s]




 73%|███████▎  | 36358/50000 [09:54<03:20, 68.01it/s]




 73%|███████▎  | 36365/50000 [09:54<03:29, 65.02it/s]




 73%|███████▎  | 36373/50000 [09:55<03:21, 67.72it/s]




 73%|███████▎  | 36380/50000 [09:55<03:28, 65.20it/s]




 73%|███████▎  | 36388/50000 [09:55<03:25, 66.17it/s]




 73%|███████▎  | 36395/50000 [09:55<03:24, 66.45it/s]




 73%|███████▎  | 36402/50000 [09:55<03:23, 66.85it/s]




 73%|███████▎  | 36410/50000 [09:55<03:14, 69.91it/s]




 73%|███████▎  | 36418/50000 [09:55<03:12, 70.55it/s]




 73%|███████▎  | 36426/50000 [09:55<03:12, 70.63it/s]




 73%|███████▎  | 36434/50000 [09:55<03:25, 66.02it/s]




 73%|███████▎  | 36442/50000 [09:56<03:18, 68.47it/s]




 73%|███████▎  | 36450/50000 [09:56<03:15, 69.39it/s]




 73%|███████▎  | 36457/50000 [09:56<03:22, 67.03it/s]




 73%|███████▎  | 36464/50000 [09:56<03:27, 65.29it/s]




 73%|███████▎  | 36471/50000 [09:56<03:24, 66.12it/s]




 73%|███████▎  | 36478/50000 [09:56<03:25, 65.78it/s]




 73%|███████▎  | 36485/50000 [09:56<03:22, 66.85it/s]




 73%|███████▎  | 36493/50000 [09:56<03:15, 69.09it/s]




 73%|███████▎  | 36500/50000 [09:56<03:27, 64.98it/s]




 73%|███████▎  | 36507/50000 [09:57<03:34, 62.94it/s]




 73%|███████▎  | 36514/50000 [09:57<03:39, 61.53it/s]




 73%|███████▎  | 36521/50000 [09:57<03:33, 63.04it/s]




 73%|███████▎  | 36529/50000 [09:57<03:26, 65.39it/s]




 73%|███████▎  | 36536/50000 [09:57<03:29, 64.37it/s]




 73%|███████▎  | 36543/50000 [09:57<03:36, 62.15it/s]




 73%|███████▎  | 36551/50000 [09:57<03:26, 65.19it/s]




 73%|███████▎  | 36559/50000 [09:57<03:19, 67.50it/s]




 73%|███████▎  | 36566/50000 [09:57<03:26, 65.06it/s]




 73%|███████▎  | 36574/50000 [09:58<03:22, 66.40it/s]




 73%|███████▎  | 36581/50000 [09:58<03:31, 63.46it/s]




 73%|███████▎  | 36589/50000 [09:58<03:20, 66.93it/s]




 73%|███████▎  | 36596/50000 [09:58<03:18, 67.68it/s]




 73%|███████▎  | 36603/50000 [09:58<03:24, 65.35it/s]




 73%|███████▎  | 36610/50000 [09:58<03:26, 64.88it/s]




 73%|███████▎  | 36618/50000 [09:58<03:18, 67.52it/s]




 73%|███████▎  | 36625/50000 [09:58<03:20, 66.65it/s]




 73%|███████▎  | 36632/50000 [09:58<03:19, 67.09it/s]




 73%|███████▎  | 36639/50000 [09:59<03:35, 62.04it/s]




 73%|███████▎  | 36646/50000 [09:59<03:33, 62.56it/s]




 73%|███████▎  | 36653/50000 [09:59<03:32, 62.76it/s]




 73%|███████▎  | 36660/50000 [09:59<03:33, 62.56it/s]




 73%|███████▎  | 36667/50000 [09:59<03:35, 61.77it/s]




 73%|███████▎  | 36674/50000 [09:59<03:30, 63.21it/s]




 73%|███████▎  | 36681/50000 [09:59<03:28, 63.99it/s]




 73%|███████▎  | 36689/50000 [09:59<03:20, 66.51it/s]




 73%|███████▎  | 36697/50000 [09:59<03:16, 67.62it/s]




 73%|███████▎  | 36704/50000 [10:00<03:17, 67.39it/s]




 73%|███████▎  | 36711/50000 [10:00<03:15, 67.99it/s]




 73%|███████▎  | 36718/50000 [10:00<03:20, 66.12it/s]




 73%|███████▎  | 36727/50000 [10:00<03:10, 69.72it/s]




 73%|███████▎  | 36735/50000 [10:00<03:12, 69.01it/s]




 73%|███████▎  | 36742/50000 [10:00<03:20, 66.03it/s]




 73%|███████▎  | 36749/50000 [10:00<03:19, 66.46it/s]




 74%|███████▎  | 36757/50000 [10:00<03:14, 68.10it/s]




 74%|███████▎  | 36764/50000 [10:00<03:14, 67.92it/s]




 74%|███████▎  | 36771/50000 [10:01<03:22, 65.33it/s]




 74%|███████▎  | 36778/50000 [10:01<03:26, 63.89it/s]




 74%|███████▎  | 36786/50000 [10:01<03:19, 66.35it/s]




 74%|███████▎  | 36794/50000 [10:01<03:14, 68.02it/s]




 74%|███████▎  | 36801/50000 [10:01<03:16, 67.08it/s]




 74%|███████▎  | 36808/50000 [10:01<03:19, 66.17it/s]




 74%|███████▎  | 36816/50000 [10:01<03:09, 69.56it/s]




 74%|███████▎  | 36824/50000 [10:01<03:18, 66.25it/s]




 74%|███████▎  | 36831/50000 [10:01<03:21, 65.41it/s]




 74%|███████▎  | 36839/50000 [10:02<03:11, 68.88it/s]




 74%|███████▎  | 36846/50000 [10:02<03:18, 66.32it/s]




 74%|███████▎  | 36854/50000 [10:02<03:12, 68.17it/s]




 74%|███████▎  | 36861/50000 [10:02<03:20, 65.49it/s]




 74%|███████▎  | 36868/50000 [10:02<03:24, 64.09it/s]




 74%|███████▍  | 36876/50000 [10:02<03:16, 66.84it/s]




 74%|███████▍  | 36883/50000 [10:02<03:21, 64.98it/s]




 74%|███████▍  | 36890/50000 [10:02<03:20, 65.25it/s]




 74%|███████▍  | 36897/50000 [10:02<03:18, 65.91it/s]




 74%|███████▍  | 36904/50000 [10:03<03:26, 63.48it/s]




 74%|███████▍  | 36912/50000 [10:03<03:16, 66.70it/s]




 74%|███████▍  | 36919/50000 [10:03<03:15, 66.94it/s]




 74%|███████▍  | 36926/50000 [10:03<03:25, 63.62it/s]




 74%|███████▍  | 36933/50000 [10:03<03:29, 62.40it/s]




 74%|███████▍  | 36940/50000 [10:03<03:26, 63.32it/s]




 74%|███████▍  | 36948/50000 [10:03<03:17, 66.09it/s]




 74%|███████▍  | 36955/50000 [10:03<03:19, 65.39it/s]




 74%|███████▍  | 36962/50000 [10:03<03:17, 66.00it/s]




 74%|███████▍  | 36970/50000 [10:04<03:10, 68.46it/s]




 74%|███████▍  | 36977/50000 [10:04<03:14, 66.80it/s]




 74%|███████▍  | 36984/50000 [10:04<03:15, 66.63it/s]




 74%|███████▍  | 36993/50000 [10:04<03:03, 70.77it/s]




 74%|███████▍  | 37001/50000 [10:04<03:16, 66.28it/s]




 74%|███████▍  | 37009/50000 [10:04<03:13, 67.29it/s]




 74%|███████▍  | 37016/50000 [10:04<03:29, 61.83it/s]




 74%|███████▍  | 37023/50000 [10:04<03:27, 62.57it/s]




 74%|███████▍  | 37030/50000 [10:04<03:25, 62.97it/s]




 74%|███████▍  | 37037/50000 [10:05<03:25, 63.13it/s]




 74%|███████▍  | 37044/50000 [10:05<03:24, 63.41it/s]




 74%|███████▍  | 37051/50000 [10:05<03:28, 62.02it/s]




 74%|███████▍  | 37058/50000 [10:05<03:21, 64.08it/s]




 74%|███████▍  | 37065/50000 [10:05<03:36, 59.75it/s]




 74%|███████▍  | 37072/50000 [10:05<03:46, 57.05it/s]




 74%|███████▍  | 37078/50000 [10:05<04:05, 52.63it/s]




 74%|███████▍  | 37084/50000 [10:05<04:27, 48.24it/s]




 74%|███████▍  | 37090/50000 [10:06<04:24, 48.73it/s]




 74%|███████▍  | 37096/50000 [10:06<04:16, 50.39it/s]




 74%|███████▍  | 37102/50000 [10:06<04:26, 48.37it/s]




 74%|███████▍  | 37107/50000 [10:06<04:39, 46.18it/s]




 74%|███████▍  | 37112/50000 [10:06<04:50, 44.29it/s]




 74%|███████▍  | 37117/50000 [10:06<04:42, 45.58it/s]




 74%|███████▍  | 37123/50000 [10:06<04:25, 48.44it/s]




 74%|███████▍  | 37129/50000 [10:06<04:11, 51.19it/s]




 74%|███████▍  | 37135/50000 [10:06<04:16, 50.10it/s]




 74%|███████▍  | 37141/50000 [10:07<04:05, 52.47it/s]




 74%|███████▍  | 37148/50000 [10:07<03:52, 55.17it/s]




 74%|███████▍  | 37154/50000 [10:07<03:53, 55.02it/s]




 74%|███████▍  | 37160/50000 [10:07<03:53, 54.92it/s]




 74%|███████▍  | 37166/50000 [10:07<04:03, 52.69it/s]




 74%|███████▍  | 37172/50000 [10:07<03:56, 54.21it/s]




 74%|███████▍  | 37179/50000 [10:07<03:44, 57.05it/s]




 74%|███████▍  | 37186/50000 [10:07<03:36, 59.22it/s]




 74%|███████▍  | 37192/50000 [10:07<03:35, 59.32it/s]




 74%|███████▍  | 37198/50000 [10:08<03:39, 58.43it/s]




 74%|███████▍  | 37204/50000 [10:08<04:15, 50.00it/s]




 74%|███████▍  | 37210/50000 [10:08<04:05, 52.11it/s]




 74%|███████▍  | 37216/50000 [10:08<04:06, 51.83it/s]




 74%|███████▍  | 37222/50000 [10:08<04:02, 52.73it/s]




 74%|███████▍  | 37228/50000 [10:08<04:07, 51.65it/s]




 74%|███████▍  | 37234/50000 [10:08<04:07, 51.58it/s]




 74%|███████▍  | 37240/50000 [10:08<04:21, 48.78it/s]




 74%|███████▍  | 37245/50000 [10:09<04:25, 48.04it/s]




 75%|███████▍  | 37252/50000 [10:09<04:05, 52.01it/s]




 75%|███████▍  | 37259/50000 [10:09<03:49, 55.45it/s]




 75%|███████▍  | 37265/50000 [10:09<03:46, 56.15it/s]




 75%|███████▍  | 37273/50000 [10:09<03:32, 59.81it/s]




 75%|███████▍  | 37280/50000 [10:09<03:26, 61.60it/s]




 75%|███████▍  | 37287/50000 [10:09<03:25, 61.91it/s]




 75%|███████▍  | 37294/50000 [10:09<03:22, 62.64it/s]




 75%|███████▍  | 37302/50000 [10:09<03:12, 65.89it/s]




 75%|███████▍  | 37309/50000 [10:10<03:19, 63.74it/s]




 75%|███████▍  | 37316/50000 [10:10<03:18, 64.02it/s]




 75%|███████▍  | 37324/50000 [10:10<03:10, 66.62it/s]




 75%|███████▍  | 37331/50000 [10:10<03:09, 66.88it/s]




 75%|███████▍  | 37338/50000 [10:10<03:07, 67.45it/s]




 75%|███████▍  | 37345/50000 [10:10<03:17, 63.95it/s]




 75%|███████▍  | 37352/50000 [10:10<03:20, 63.05it/s]




 75%|███████▍  | 37359/50000 [10:10<03:17, 64.14it/s]




 75%|███████▍  | 37366/50000 [10:10<03:21, 62.67it/s]




 75%|███████▍  | 37373/50000 [10:11<03:20, 63.01it/s]




 75%|███████▍  | 37381/50000 [10:11<03:10, 66.17it/s]




 75%|███████▍  | 37388/50000 [10:11<03:14, 64.90it/s]




 75%|███████▍  | 37395/50000 [10:11<03:17, 63.69it/s]




 75%|███████▍  | 37402/50000 [10:11<03:53, 53.92it/s]




 75%|███████▍  | 37408/50000 [10:11<04:11, 50.15it/s]




 75%|███████▍  | 37415/50000 [10:11<03:50, 54.53it/s]




 75%|███████▍  | 37421/50000 [10:11<03:47, 55.33it/s]




 75%|███████▍  | 37429/50000 [10:12<03:34, 58.73it/s]




 75%|███████▍  | 37436/50000 [10:12<03:31, 59.40it/s]




 75%|███████▍  | 37443/50000 [10:12<03:26, 60.82it/s]




 75%|███████▍  | 37451/50000 [10:12<03:17, 63.55it/s]




 75%|███████▍  | 37458/50000 [10:12<03:14, 64.32it/s]




 75%|███████▍  | 37465/50000 [10:12<03:27, 60.52it/s]




 75%|███████▍  | 37472/50000 [10:12<03:27, 60.49it/s]




 75%|███████▍  | 37479/50000 [10:12<03:32, 59.04it/s]




 75%|███████▍  | 37486/50000 [10:12<03:30, 59.36it/s]




 75%|███████▍  | 37493/50000 [10:13<03:24, 61.11it/s]




 75%|███████▌  | 37501/50000 [10:13<03:10, 65.49it/s]




 75%|███████▌  | 37508/50000 [10:13<03:15, 63.91it/s]




 75%|███████▌  | 37516/50000 [10:13<03:04, 67.65it/s]




 75%|███████▌  | 37524/50000 [10:13<03:00, 69.15it/s]




 75%|███████▌  | 37532/50000 [10:13<02:58, 69.69it/s]




 75%|███████▌  | 37540/50000 [10:13<03:02, 68.46it/s]




 75%|███████▌  | 37547/50000 [10:13<03:08, 66.05it/s]




 75%|███████▌  | 37554/50000 [10:13<03:12, 64.82it/s]




 75%|███████▌  | 37561/50000 [10:14<03:10, 65.23it/s]




 75%|███████▌  | 37568/50000 [10:14<03:17, 62.92it/s]




 75%|███████▌  | 37575/50000 [10:14<03:14, 63.87it/s]




 75%|███████▌  | 37583/50000 [10:14<03:05, 66.92it/s]




 75%|███████▌  | 37590/50000 [10:14<03:04, 67.28it/s]




 75%|███████▌  | 37597/50000 [10:14<03:04, 67.34it/s]




 75%|███████▌  | 37604/50000 [10:14<03:05, 66.81it/s]




 75%|███████▌  | 37611/50000 [10:14<03:04, 67.20it/s]




 75%|███████▌  | 37618/50000 [10:14<03:04, 67.29it/s]




 75%|███████▌  | 37625/50000 [10:14<03:06, 66.20it/s]




 75%|███████▌  | 37632/50000 [10:15<03:14, 63.51it/s]




 75%|███████▌  | 37639/50000 [10:15<03:11, 64.65it/s]




 75%|███████▌  | 37646/50000 [10:15<03:08, 65.48it/s]




 75%|███████▌  | 37653/50000 [10:15<03:10, 64.79it/s]




 75%|███████▌  | 37660/50000 [10:15<03:11, 64.31it/s]




 75%|███████▌  | 37667/50000 [10:15<03:11, 64.34it/s]




 75%|███████▌  | 37675/50000 [10:15<03:04, 66.86it/s]




 75%|███████▌  | 37682/50000 [10:15<03:03, 66.95it/s]




 75%|███████▌  | 37689/50000 [10:15<03:04, 66.80it/s]




 75%|███████▌  | 37696/50000 [10:16<03:06, 65.87it/s]




 75%|███████▌  | 37703/50000 [10:16<03:12, 63.99it/s]




 75%|███████▌  | 37711/50000 [10:16<03:03, 66.83it/s]




 75%|███████▌  | 37718/50000 [10:16<03:07, 65.52it/s]




 75%|███████▌  | 37725/50000 [10:16<03:08, 65.00it/s]




 75%|███████▌  | 37733/50000 [10:16<02:58, 68.57it/s]




 75%|███████▌  | 37740/50000 [10:16<03:01, 67.65it/s]




 75%|███████▌  | 37748/50000 [10:16<02:58, 68.79it/s]




 76%|███████▌  | 37755/50000 [10:16<03:04, 66.45it/s]




 76%|███████▌  | 37762/50000 [10:17<03:13, 63.33it/s]




 76%|███████▌  | 37769/50000 [10:17<03:12, 63.47it/s]




 76%|███████▌  | 37776/50000 [10:17<03:19, 61.24it/s]




 76%|███████▌  | 37783/50000 [10:17<03:21, 60.62it/s]




 76%|███████▌  | 37790/50000 [10:17<03:19, 61.14it/s]




 76%|███████▌  | 37798/50000 [10:17<03:08, 64.88it/s]




 76%|███████▌  | 37805/50000 [10:17<03:07, 64.99it/s]




 76%|███████▌  | 37812/50000 [10:17<03:04, 65.90it/s]




 76%|███████▌  | 37821/50000 [10:17<02:54, 69.87it/s]




 76%|███████▌  | 37829/50000 [10:18<02:57, 68.43it/s]




 76%|███████▌  | 37836/50000 [10:18<03:00, 67.55it/s]




 76%|███████▌  | 37843/50000 [10:18<03:00, 67.34it/s]




 76%|███████▌  | 37850/50000 [10:18<03:01, 67.00it/s]




 76%|███████▌  | 37859/50000 [10:18<02:51, 70.73it/s]




 76%|███████▌  | 37867/50000 [10:18<02:55, 69.16it/s]




 76%|███████▌  | 37874/50000 [10:18<03:00, 67.27it/s]




 76%|███████▌  | 37882/50000 [10:18<02:55, 69.23it/s]




 76%|███████▌  | 37889/50000 [10:18<02:59, 67.31it/s]




 76%|███████▌  | 37896/50000 [10:19<03:07, 64.66it/s]




 76%|███████▌  | 37903/50000 [10:19<03:06, 64.76it/s]




 76%|███████▌  | 37910/50000 [10:19<03:13, 62.41it/s]




 76%|███████▌  | 37917/50000 [10:19<03:09, 63.68it/s]




 76%|███████▌  | 37924/50000 [10:19<03:05, 64.95it/s]




 76%|███████▌  | 37931/50000 [10:19<03:05, 65.05it/s]




 76%|███████▌  | 37939/50000 [10:19<03:02, 66.19it/s]




 76%|███████▌  | 37946/50000 [10:19<03:04, 65.45it/s]




 76%|███████▌  | 37953/50000 [10:19<03:00, 66.62it/s]




 76%|███████▌  | 37960/50000 [10:20<02:59, 67.07it/s]




 76%|███████▌  | 37967/50000 [10:20<02:59, 67.19it/s]




 76%|███████▌  | 37974/50000 [10:20<03:06, 64.50it/s]




 76%|███████▌  | 37981/50000 [10:20<03:17, 60.94it/s]




 76%|███████▌  | 37988/50000 [10:20<03:11, 62.60it/s]




 76%|███████▌  | 37995/50000 [10:20<03:10, 62.96it/s]




 76%|███████▌  | 38002/50000 [10:20<03:11, 62.61it/s]




 76%|███████▌  | 38010/50000 [10:20<02:59, 66.87it/s]




 76%|███████▌  | 38018/50000 [10:20<02:56, 67.96it/s]




 76%|███████▌  | 38025/50000 [10:21<03:00, 66.47it/s]




 76%|███████▌  | 38034/50000 [10:21<02:48, 70.89it/s]




 76%|███████▌  | 38042/50000 [10:21<02:52, 69.52it/s]




 76%|███████▌  | 38050/50000 [10:21<02:55, 68.25it/s]




 76%|███████▌  | 38057/50000 [10:21<03:24, 58.28it/s]




 76%|███████▌  | 38064/50000 [10:21<03:37, 54.79it/s]




 76%|███████▌  | 38070/50000 [10:21<03:39, 54.31it/s]




 76%|███████▌  | 38076/50000 [10:21<03:43, 53.41it/s]




 76%|███████▌  | 38082/50000 [10:22<04:04, 48.81it/s]




 76%|███████▌  | 38088/50000 [10:22<03:58, 49.94it/s]




 76%|███████▌  | 38094/50000 [10:22<03:51, 51.54it/s]




 76%|███████▌  | 38100/50000 [10:22<03:49, 51.83it/s]




 76%|███████▌  | 38106/50000 [10:22<03:49, 51.77it/s]




 76%|███████▌  | 38112/50000 [10:22<03:45, 52.81it/s]




 76%|███████▌  | 38118/50000 [10:22<03:44, 52.94it/s]




 76%|███████▌  | 38124/50000 [10:22<03:43, 53.17it/s]




 76%|███████▋  | 38130/50000 [10:23<03:40, 53.90it/s]




 76%|███████▋  | 38137/50000 [10:23<03:28, 56.81it/s]




 76%|███████▋  | 38145/50000 [10:23<03:12, 61.69it/s]




 76%|███████▋  | 38152/50000 [10:23<03:10, 62.31it/s]




 76%|███████▋  | 38159/50000 [10:23<03:11, 61.76it/s]




 76%|███████▋  | 38167/50000 [10:23<03:02, 64.73it/s]




 76%|███████▋  | 38175/50000 [10:23<02:55, 67.32it/s]




 76%|███████▋  | 38182/50000 [10:23<02:53, 67.96it/s]




 76%|███████▋  | 38189/50000 [10:23<02:59, 65.91it/s]




 76%|███████▋  | 38196/50000 [10:23<02:59, 65.91it/s]




 76%|███████▋  | 38203/50000 [10:24<03:15, 60.38it/s]




 76%|███████▋  | 38211/50000 [10:24<03:03, 64.32it/s]




 76%|███████▋  | 38218/50000 [10:24<03:00, 65.42it/s]




 76%|███████▋  | 38225/50000 [10:24<03:30, 56.06it/s]




 76%|███████▋  | 38231/50000 [10:24<03:45, 52.17it/s]




 76%|███████▋  | 38237/50000 [10:24<03:51, 50.89it/s]




 76%|███████▋  | 38243/50000 [10:24<03:40, 53.21it/s]




 76%|███████▋  | 38249/50000 [10:24<03:34, 54.89it/s]




 77%|███████▋  | 38255/50000 [10:25<03:31, 55.51it/s]




 77%|███████▋  | 38263/50000 [10:25<03:17, 59.40it/s]




 77%|███████▋  | 38270/50000 [10:25<03:18, 59.12it/s]




 77%|███████▋  | 38277/50000 [10:25<03:17, 59.23it/s]




 77%|███████▋  | 38285/50000 [10:25<03:06, 62.76it/s]




 77%|███████▋  | 38293/50000 [10:25<02:57, 65.82it/s]




 77%|███████▋  | 38300/50000 [10:25<03:00, 64.84it/s]




 77%|███████▋  | 38307/50000 [10:25<02:59, 65.25it/s]




 77%|███████▋  | 38314/50000 [10:25<02:56, 66.18it/s]




 77%|███████▋  | 38321/50000 [10:26<02:59, 65.09it/s]




 77%|███████▋  | 38328/50000 [10:26<03:02, 63.82it/s]




 77%|███████▋  | 38336/50000 [10:26<02:55, 66.63it/s]




 77%|███████▋  | 38344/50000 [10:26<02:49, 68.93it/s]




 77%|███████▋  | 38351/50000 [10:26<02:53, 67.05it/s]




 77%|███████▋  | 38358/50000 [10:26<02:51, 67.86it/s]




 77%|███████▋  | 38365/50000 [10:26<02:53, 67.16it/s]




 77%|███████▋  | 38373/50000 [10:26<02:45, 70.24it/s]




 77%|███████▋  | 38381/50000 [10:26<02:41, 71.79it/s]




 77%|███████▋  | 38389/50000 [10:27<02:55, 66.23it/s]




 77%|███████▋  | 38396/50000 [10:27<03:03, 63.18it/s]




 77%|███████▋  | 38403/50000 [10:27<03:01, 64.06it/s]




 77%|███████▋  | 38410/50000 [10:27<03:01, 63.81it/s]




 77%|███████▋  | 38417/50000 [10:27<02:58, 65.05it/s]




 77%|███████▋  | 38424/50000 [10:27<02:57, 65.21it/s]




 77%|███████▋  | 38431/50000 [10:27<02:59, 64.61it/s]




 77%|███████▋  | 38439/50000 [10:27<02:53, 66.51it/s]




 77%|███████▋  | 38446/50000 [10:27<02:51, 67.51it/s]




 77%|███████▋  | 38453/50000 [10:28<02:57, 65.15it/s]




 77%|███████▋  | 38460/50000 [10:28<02:58, 64.83it/s]




 77%|███████▋  | 38467/50000 [10:28<02:54, 66.16it/s]




 77%|███████▋  | 38474/50000 [10:28<02:56, 65.25it/s]




 77%|███████▋  | 38481/50000 [10:28<02:55, 65.72it/s]




 77%|███████▋  | 38488/50000 [10:28<02:52, 66.81it/s]




 77%|███████▋  | 38496/50000 [10:28<02:49, 68.01it/s]




 77%|███████▋  | 38503/50000 [10:28<02:56, 65.02it/s]




 77%|███████▋  | 38510/50000 [10:28<03:05, 62.08it/s]




 77%|███████▋  | 38517/50000 [10:29<03:12, 59.71it/s]




 77%|███████▋  | 38524/50000 [10:29<03:04, 62.35it/s]




 77%|███████▋  | 38531/50000 [10:29<03:00, 63.46it/s]




 77%|███████▋  | 38539/50000 [10:29<02:52, 66.36it/s]




 77%|███████▋  | 38546/50000 [10:29<03:07, 61.12it/s]




 77%|███████▋  | 38553/50000 [10:29<03:01, 63.15it/s]




 77%|███████▋  | 38562/50000 [10:29<02:48, 67.69it/s]




 77%|███████▋  | 38570/50000 [10:29<02:43, 69.81it/s]




 77%|███████▋  | 38578/50000 [10:29<02:48, 67.67it/s]




 77%|███████▋  | 38585/50000 [10:30<02:56, 64.63it/s]




 77%|███████▋  | 38593/50000 [10:30<02:50, 67.08it/s]




 77%|███████▋  | 38600/50000 [10:30<02:56, 64.42it/s]




 77%|███████▋  | 38607/50000 [10:30<03:00, 63.20it/s]




 77%|███████▋  | 38614/50000 [10:30<03:02, 62.37it/s]




 77%|███████▋  | 38622/50000 [10:30<02:54, 65.35it/s]




 77%|███████▋  | 38629/50000 [10:30<02:54, 65.16it/s]




 77%|███████▋  | 38636/50000 [10:30<02:51, 66.40it/s]




 77%|███████▋  | 38645/50000 [10:30<02:41, 70.33it/s]




 77%|███████▋  | 38653/50000 [10:31<02:40, 70.90it/s]




 77%|███████▋  | 38661/50000 [10:31<02:45, 68.39it/s]




 77%|███████▋  | 38668/50000 [10:31<02:51, 65.89it/s]




 77%|███████▋  | 38675/50000 [10:31<03:01, 62.55it/s]




 77%|███████▋  | 38682/50000 [10:31<02:57, 63.78it/s]




 77%|███████▋  | 38690/50000 [10:31<02:51, 65.94it/s]




 77%|███████▋  | 38697/50000 [10:31<02:55, 64.39it/s]




 77%|███████▋  | 38704/50000 [10:31<02:57, 63.69it/s]




 77%|███████▋  | 38713/50000 [10:31<02:48, 67.05it/s]




 77%|███████▋  | 38721/50000 [10:32<02:41, 69.97it/s]




 77%|███████▋  | 38729/50000 [10:32<02:45, 67.95it/s]




 77%|███████▋  | 38736/50000 [10:32<02:53, 64.81it/s]




 77%|███████▋  | 38743/50000 [10:32<02:53, 64.86it/s]




 78%|███████▊  | 38751/50000 [10:32<02:48, 66.92it/s]




 78%|███████▊  | 38758/50000 [10:32<02:56, 63.53it/s]




 78%|███████▊  | 38765/50000 [10:32<03:02, 61.52it/s]




 78%|███████▊  | 38772/50000 [10:32<03:01, 62.03it/s]




 78%|███████▊  | 38780/50000 [10:33<02:52, 65.10it/s]




 78%|███████▊  | 38787/50000 [10:33<02:52, 65.06it/s]




 78%|███████▊  | 38794/50000 [10:33<02:51, 65.22it/s]




 78%|███████▊  | 38801/50000 [10:33<03:06, 59.97it/s]




 78%|███████▊  | 38808/50000 [10:33<03:16, 57.05it/s]




 78%|███████▊  | 38814/50000 [10:33<03:23, 54.94it/s]




 78%|███████▊  | 38821/50000 [10:33<03:12, 58.04it/s]




 78%|███████▊  | 38827/50000 [10:33<03:21, 55.41it/s]




 78%|███████▊  | 38835/50000 [10:33<03:07, 59.58it/s]




 78%|███████▊  | 38842/50000 [10:34<03:16, 56.80it/s]




 78%|███████▊  | 38848/50000 [10:34<03:19, 55.76it/s]




 78%|███████▊  | 38854/50000 [10:34<03:23, 54.90it/s]




 78%|███████▊  | 38860/50000 [10:34<03:18, 56.07it/s]




 78%|███████▊  | 38867/50000 [10:34<03:17, 56.50it/s]




 78%|███████▊  | 38873/50000 [10:34<03:17, 56.41it/s]




 78%|███████▊  | 38879/50000 [10:34<03:23, 54.67it/s]




 78%|███████▊  | 38885/50000 [10:34<03:22, 54.97it/s]




 78%|███████▊  | 38891/50000 [10:34<03:24, 54.44it/s]




 78%|███████▊  | 38897/50000 [10:35<03:25, 54.08it/s]




 78%|███████▊  | 38903/50000 [10:35<03:19, 55.61it/s]




 78%|███████▊  | 38910/50000 [10:35<03:08, 58.72it/s]




 78%|███████▊  | 38917/50000 [10:35<03:06, 59.54it/s]




 78%|███████▊  | 38923/50000 [10:35<03:09, 58.34it/s]




 78%|███████▊  | 38929/50000 [10:35<03:08, 58.70it/s]




 78%|███████▊  | 38936/50000 [10:35<03:02, 60.55it/s]




 78%|███████▊  | 38943/50000 [10:35<03:13, 57.13it/s]




 78%|███████▊  | 38950/50000 [10:35<03:07, 58.98it/s]




 78%|███████▊  | 38956/50000 [10:36<03:08, 58.64it/s]




 78%|███████▊  | 38963/50000 [10:36<03:06, 59.19it/s]




 78%|███████▊  | 38969/50000 [10:36<03:20, 55.07it/s]




 78%|███████▊  | 38975/50000 [10:36<03:19, 55.26it/s]




 78%|███████▊  | 38982/50000 [10:36<03:09, 58.29it/s]




 78%|███████▊  | 38989/50000 [10:36<03:04, 59.84it/s]




 78%|███████▊  | 38996/50000 [10:36<02:59, 61.43it/s]




 78%|███████▊  | 39004/50000 [10:36<02:50, 64.32it/s]




 78%|███████▊  | 39011/50000 [10:36<02:54, 63.13it/s]




 78%|███████▊  | 39019/50000 [10:37<02:46, 65.94it/s]




 78%|███████▊  | 39026/50000 [10:37<02:45, 66.21it/s]




 78%|███████▊  | 39033/50000 [10:37<02:50, 64.47it/s]




 78%|███████▊  | 39040/50000 [10:37<02:55, 62.47it/s]




 78%|███████▊  | 39047/50000 [10:37<03:03, 59.73it/s]




 78%|███████▊  | 39054/50000 [10:37<03:18, 55.16it/s]




 78%|███████▊  | 39060/50000 [10:37<03:31, 51.62it/s]




 78%|███████▊  | 39066/50000 [10:37<03:26, 52.92it/s]




 78%|███████▊  | 39072/50000 [10:38<03:29, 52.19it/s]




 78%|███████▊  | 39078/50000 [10:38<03:46, 48.21it/s]




 78%|███████▊  | 39083/50000 [10:38<03:46, 48.21it/s]




 78%|███████▊  | 39088/50000 [10:38<04:04, 44.54it/s]




 78%|███████▊  | 39094/50000 [10:38<03:55, 46.24it/s]




 78%|███████▊  | 39099/50000 [10:38<03:56, 46.04it/s]




 78%|███████▊  | 39104/50000 [10:38<03:58, 45.77it/s]




 78%|███████▊  | 39110/50000 [10:38<03:47, 47.78it/s]




 78%|███████▊  | 39115/50000 [10:39<03:56, 45.93it/s]




 78%|███████▊  | 39121/50000 [10:39<03:47, 47.91it/s]




 78%|███████▊  | 39129/50000 [10:39<03:22, 53.62it/s]




 78%|███████▊  | 39136/50000 [10:39<03:11, 56.59it/s]




 78%|███████▊  | 39143/50000 [10:39<03:01, 59.93it/s]




 78%|███████▊  | 39150/50000 [10:39<02:59, 60.56it/s]




 78%|███████▊  | 39157/50000 [10:39<03:04, 58.87it/s]




 78%|███████▊  | 39164/50000 [10:39<02:58, 60.74it/s]




 78%|███████▊  | 39171/50000 [10:39<02:52, 62.95it/s]




 78%|███████▊  | 39178/50000 [10:40<02:56, 61.38it/s]




 78%|███████▊  | 39185/50000 [10:40<03:00, 59.87it/s]




 78%|███████▊  | 39192/50000 [10:40<03:14, 55.60it/s]




 78%|███████▊  | 39200/50000 [10:40<03:01, 59.67it/s]




 78%|███████▊  | 39207/50000 [10:40<03:02, 59.23it/s]




 78%|███████▊  | 39214/50000 [10:40<03:16, 54.99it/s]




 78%|███████▊  | 39220/50000 [10:40<03:46, 47.60it/s]




 78%|███████▊  | 39226/50000 [10:40<03:59, 45.07it/s]




 78%|███████▊  | 39232/50000 [10:41<03:42, 48.39it/s]




 78%|███████▊  | 39239/50000 [10:41<03:23, 52.88it/s]




 78%|███████▊  | 39247/50000 [10:41<03:06, 57.75it/s]




 79%|███████▊  | 39254/50000 [10:41<02:59, 59.90it/s]




 79%|███████▊  | 39261/50000 [10:41<03:01, 59.09it/s]




 79%|███████▊  | 39268/50000 [10:41<02:54, 61.38it/s]




 79%|███████▊  | 39275/50000 [10:41<02:58, 60.17it/s]




 79%|███████▊  | 39282/50000 [10:41<02:58, 59.96it/s]




 79%|███████▊  | 39289/50000 [10:41<02:55, 60.99it/s]




 79%|███████▊  | 39296/50000 [10:42<02:51, 62.47it/s]




 79%|███████▊  | 39304/50000 [10:42<02:43, 65.43it/s]




 79%|███████▊  | 39311/50000 [10:42<02:43, 65.48it/s]




 79%|███████▊  | 39319/50000 [10:42<02:38, 67.38it/s]




 79%|███████▊  | 39327/50000 [10:42<02:33, 69.31it/s]




 79%|███████▊  | 39334/50000 [10:42<02:37, 67.76it/s]




 79%|███████▊  | 39341/50000 [10:42<02:46, 64.15it/s]




 79%|███████▊  | 39350/50000 [10:42<02:33, 69.59it/s]




 79%|███████▊  | 39358/50000 [10:42<02:32, 69.82it/s]




 79%|███████▊  | 39366/50000 [10:43<02:38, 67.00it/s]




 79%|███████▊  | 39373/50000 [10:43<02:43, 65.09it/s]




 79%|███████▉  | 39380/50000 [10:43<02:48, 63.12it/s]




 79%|███████▉  | 39388/50000 [10:43<02:43, 64.77it/s]




 79%|███████▉  | 39395/50000 [10:43<02:48, 63.00it/s]




 79%|███████▉  | 39402/50000 [10:43<02:43, 64.82it/s]




 79%|███████▉  | 39409/50000 [10:43<02:42, 65.05it/s]




 79%|███████▉  | 39416/50000 [10:43<02:43, 64.67it/s]




 79%|███████▉  | 39423/50000 [10:44<02:43, 64.77it/s]




 79%|███████▉  | 39430/50000 [10:44<02:51, 61.75it/s]




 79%|███████▉  | 39437/50000 [10:44<02:45, 63.71it/s]




 79%|███████▉  | 39445/50000 [10:44<02:39, 66.06it/s]




 79%|███████▉  | 39452/50000 [10:44<02:37, 66.76it/s]




 79%|███████▉  | 39459/50000 [10:44<02:42, 64.68it/s]




 79%|███████▉  | 39466/50000 [10:44<02:43, 64.41it/s]




 79%|███████▉  | 39473/50000 [10:44<02:42, 64.66it/s]




 79%|███████▉  | 39480/50000 [10:44<02:39, 65.85it/s]




 79%|███████▉  | 39487/50000 [10:44<02:39, 65.77it/s]




 79%|███████▉  | 39494/50000 [10:45<02:52, 60.98it/s]




 79%|███████▉  | 39501/50000 [10:45<02:50, 61.55it/s]




 79%|███████▉  | 39509/50000 [10:45<02:43, 64.01it/s]




 79%|███████▉  | 39517/50000 [10:45<02:34, 67.79it/s]




 79%|███████▉  | 39524/50000 [10:45<02:42, 64.53it/s]




 79%|███████▉  | 39531/50000 [10:45<03:05, 56.54it/s]




 79%|███████▉  | 39539/50000 [10:45<02:50, 61.18it/s]




 79%|███████▉  | 39546/50000 [10:45<02:59, 58.10it/s]




 79%|███████▉  | 39553/50000 [10:46<03:20, 52.12it/s]




 79%|███████▉  | 39560/50000 [10:46<03:06, 56.08it/s]




 79%|███████▉  | 39567/50000 [10:46<02:56, 59.07it/s]




 79%|███████▉  | 39574/50000 [10:46<02:59, 58.16it/s]




 79%|███████▉  | 39580/50000 [10:46<03:03, 56.91it/s]




 79%|███████▉  | 39588/50000 [10:46<02:50, 61.07it/s]




 79%|███████▉  | 39596/50000 [10:46<02:41, 64.58it/s]




 79%|███████▉  | 39603/50000 [10:46<02:40, 64.79it/s]




 79%|███████▉  | 39610/50000 [10:47<02:57, 58.38it/s]




 79%|███████▉  | 39617/50000 [10:47<02:55, 59.30it/s]




 79%|███████▉  | 39624/50000 [10:47<02:52, 60.27it/s]




 79%|███████▉  | 39631/50000 [10:47<02:56, 58.82it/s]




 79%|███████▉  | 39637/50000 [10:47<03:02, 56.87it/s]




 79%|███████▉  | 39644/50000 [10:47<02:53, 59.53it/s]




 79%|███████▉  | 39651/50000 [10:47<02:49, 61.07it/s]




 79%|███████▉  | 39658/50000 [10:47<02:48, 61.22it/s]




 79%|███████▉  | 39665/50000 [10:47<02:54, 59.36it/s]




 79%|███████▉  | 39671/50000 [10:48<02:58, 57.88it/s]




 79%|███████▉  | 39677/50000 [10:48<02:57, 58.03it/s]




 79%|███████▉  | 39683/50000 [10:48<03:09, 54.55it/s]




 79%|███████▉  | 39690/50000 [10:48<02:59, 57.52it/s]




 79%|███████▉  | 39697/50000 [10:48<02:53, 59.50it/s]




 79%|███████▉  | 39704/50000 [10:48<02:45, 62.09it/s]




 79%|███████▉  | 39711/50000 [10:48<02:46, 61.69it/s]




 79%|███████▉  | 39718/50000 [10:48<02:47, 61.33it/s]




 79%|███████▉  | 39725/50000 [10:48<02:57, 57.76it/s]




 79%|███████▉  | 39732/50000 [10:49<02:54, 58.71it/s]




 79%|███████▉  | 39738/50000 [10:49<03:00, 56.95it/s]




 79%|███████▉  | 39744/50000 [10:49<03:00, 56.73it/s]




 80%|███████▉  | 39750/50000 [10:49<03:03, 55.95it/s]




 80%|███████▉  | 39757/50000 [10:49<02:58, 57.52it/s]




 80%|███████▉  | 39764/50000 [10:49<02:51, 59.80it/s]




 80%|███████▉  | 39771/50000 [10:49<02:53, 58.95it/s]




 80%|███████▉  | 39777/50000 [10:49<02:55, 58.27it/s]




 80%|███████▉  | 39783/50000 [10:49<02:59, 56.83it/s]




 80%|███████▉  | 39789/50000 [10:50<03:02, 56.09it/s]




 80%|███████▉  | 39795/50000 [10:50<03:01, 56.21it/s]




 80%|███████▉  | 39801/50000 [10:50<03:05, 54.94it/s]




 80%|███████▉  | 39807/50000 [10:50<03:01, 56.01it/s]




 80%|███████▉  | 39813/50000 [10:50<03:12, 53.03it/s]




 80%|███████▉  | 39819/50000 [10:50<03:14, 52.26it/s]




 80%|███████▉  | 39825/50000 [10:50<03:09, 53.82it/s]




 80%|███████▉  | 39831/50000 [10:50<03:15, 52.11it/s]




 80%|███████▉  | 39837/50000 [10:51<03:16, 51.77it/s]




 80%|███████▉  | 39843/50000 [10:51<03:08, 53.88it/s]




 80%|███████▉  | 39850/50000 [10:51<02:55, 57.78it/s]




 80%|███████▉  | 39856/50000 [10:51<03:12, 52.78it/s]




 80%|███████▉  | 39862/50000 [10:51<03:05, 54.72it/s]




 80%|███████▉  | 39869/50000 [10:51<02:57, 57.02it/s]




 80%|███████▉  | 39875/50000 [10:51<03:06, 54.32it/s]




 80%|███████▉  | 39881/50000 [10:51<03:03, 55.03it/s]




 80%|███████▉  | 39887/50000 [10:51<03:09, 53.25it/s]




 80%|███████▉  | 39893/50000 [10:52<03:03, 54.94it/s]




 80%|███████▉  | 39899/50000 [10:52<03:01, 55.70it/s]




 80%|███████▉  | 39905/50000 [10:52<03:10, 53.05it/s]




 80%|███████▉  | 39911/50000 [10:52<03:07, 53.82it/s]




 80%|███████▉  | 39917/50000 [10:52<03:06, 54.08it/s]




 80%|███████▉  | 39925/50000 [10:52<02:51, 58.88it/s]




 80%|███████▉  | 39932/50000 [10:52<02:50, 59.06it/s]




 80%|███████▉  | 39939/50000 [10:52<02:47, 60.10it/s]




 80%|███████▉  | 39946/50000 [10:52<02:43, 61.65it/s]




 80%|███████▉  | 39953/50000 [10:53<02:39, 62.95it/s]




 80%|███████▉  | 39960/50000 [10:53<02:37, 63.92it/s]




 80%|███████▉  | 39967/50000 [10:53<02:35, 64.41it/s]




 80%|███████▉  | 39974/50000 [10:53<02:39, 62.68it/s]




 80%|███████▉  | 39982/50000 [10:53<02:33, 65.12it/s]




 80%|███████▉  | 39989/50000 [10:53<02:59, 55.77it/s]




 80%|███████▉  | 39995/50000 [10:53<03:24, 49.00it/s]




 80%|████████  | 40001/50000 [10:53<03:36, 46.10it/s]




 80%|████████  | 40006/50000 [10:54<04:00, 41.60it/s]




 80%|████████  | 40011/50000 [10:54<03:54, 42.61it/s]




 80%|████████  | 40017/50000 [10:54<03:46, 44.13it/s]




 80%|████████  | 40022/50000 [10:54<03:42, 44.91it/s]




 80%|████████  | 40027/50000 [10:54<03:43, 44.63it/s]




 80%|████████  | 40033/50000 [10:54<03:38, 45.52it/s]




 80%|████████  | 40038/50000 [10:54<03:35, 46.23it/s]




 80%|████████  | 40043/50000 [10:54<03:53, 42.62it/s]




 80%|████████  | 40048/50000 [10:55<04:05, 40.52it/s]




 80%|████████  | 40053/50000 [10:55<04:00, 41.39it/s]




 80%|████████  | 40060/50000 [10:55<03:35, 46.09it/s]




 80%|████████  | 40067/50000 [10:55<03:17, 50.38it/s]




 80%|████████  | 40074/50000 [10:55<03:07, 52.93it/s]




 80%|████████  | 40080/50000 [10:55<03:07, 52.95it/s]




 80%|████████  | 40086/50000 [10:55<03:00, 54.85it/s]




 80%|████████  | 40092/50000 [10:55<03:00, 54.80it/s]




 80%|████████  | 40099/50000 [10:55<02:51, 57.79it/s]




 80%|████████  | 40105/50000 [10:56<02:55, 56.34it/s]




 80%|████████  | 40112/50000 [10:56<02:50, 57.96it/s]




 80%|████████  | 40119/50000 [10:56<02:45, 59.75it/s]




 80%|████████  | 40126/50000 [10:56<03:06, 52.82it/s]




 80%|████████  | 40132/50000 [10:56<03:18, 49.75it/s]




 80%|████████  | 40138/50000 [10:56<03:40, 44.65it/s]




 80%|████████  | 40143/50000 [10:56<03:37, 45.41it/s]




 80%|████████  | 40148/50000 [10:56<03:48, 43.06it/s]




 80%|████████  | 40153/50000 [10:57<03:54, 41.92it/s]




 80%|████████  | 40158/50000 [10:57<03:52, 42.41it/s]




 80%|████████  | 40163/50000 [10:57<03:44, 43.76it/s]




 80%|████████  | 40171/50000 [10:57<03:17, 49.73it/s]




 80%|████████  | 40178/50000 [10:57<03:00, 54.30it/s]




 80%|████████  | 40184/50000 [10:57<03:11, 51.22it/s]




 80%|████████  | 40190/50000 [10:57<03:10, 51.41it/s]




 80%|████████  | 40197/50000 [10:57<02:59, 54.72it/s]




 80%|████████  | 40204/50000 [10:57<02:49, 57.73it/s]




 80%|████████  | 40210/50000 [10:58<02:48, 57.93it/s]




 80%|████████  | 40217/50000 [10:58<02:46, 58.83it/s]




 80%|████████  | 40224/50000 [10:58<02:39, 61.18it/s]




 80%|████████  | 40231/50000 [10:58<02:35, 62.78it/s]




 80%|████████  | 40238/50000 [10:58<02:37, 62.11it/s]




 80%|████████  | 40245/50000 [10:58<02:41, 60.35it/s]




 81%|████████  | 40252/50000 [10:58<03:00, 53.86it/s]




 81%|████████  | 40258/50000 [10:58<03:07, 51.87it/s]




 81%|████████  | 40266/50000 [10:59<02:52, 56.31it/s]




 81%|████████  | 40273/50000 [10:59<02:45, 58.81it/s]




 81%|████████  | 40280/50000 [10:59<02:37, 61.65it/s]




 81%|████████  | 40287/50000 [10:59<02:33, 63.46it/s]




 81%|████████  | 40294/50000 [10:59<02:28, 65.16it/s]




 81%|████████  | 40301/50000 [10:59<02:37, 61.51it/s]




 81%|████████  | 40308/50000 [10:59<02:37, 61.53it/s]




 81%|████████  | 40315/50000 [10:59<02:37, 61.38it/s]




 81%|████████  | 40323/50000 [10:59<02:30, 64.51it/s]




 81%|████████  | 40330/50000 [11:00<02:29, 64.83it/s]




 81%|████████  | 40337/50000 [11:00<02:35, 62.12it/s]




 81%|████████  | 40346/50000 [11:00<02:25, 66.41it/s]




 81%|████████  | 40353/50000 [11:00<02:25, 66.22it/s]




 81%|████████  | 40360/50000 [11:00<02:25, 66.12it/s]




 81%|████████  | 40367/50000 [11:00<02:25, 66.15it/s]




 81%|████████  | 40374/50000 [11:00<02:30, 63.82it/s]




 81%|████████  | 40381/50000 [11:00<02:31, 63.30it/s]




 81%|████████  | 40388/50000 [11:00<02:28, 64.68it/s]




 81%|████████  | 40395/50000 [11:01<02:31, 63.54it/s]




 81%|████████  | 40402/50000 [11:01<02:31, 63.28it/s]




 81%|████████  | 40409/50000 [11:01<02:27, 64.84it/s]




 81%|████████  | 40416/50000 [11:01<02:30, 63.82it/s]




 81%|████████  | 40424/50000 [11:01<02:24, 66.14it/s]




 81%|████████  | 40431/50000 [11:01<02:24, 66.05it/s]




 81%|████████  | 40438/50000 [11:01<02:36, 60.93it/s]




 81%|████████  | 40445/50000 [11:01<02:37, 60.80it/s]




 81%|████████  | 40452/50000 [11:01<02:32, 62.41it/s]




 81%|████████  | 40459/50000 [11:02<02:40, 59.54it/s]




 81%|████████  | 40466/50000 [11:02<02:33, 62.21it/s]




 81%|████████  | 40473/50000 [11:02<02:36, 60.73it/s]




 81%|████████  | 40480/50000 [11:02<02:41, 58.98it/s]




 81%|████████  | 40486/50000 [11:02<02:46, 57.13it/s]




 81%|████████  | 40492/50000 [11:02<02:47, 56.86it/s]




 81%|████████  | 40499/50000 [11:02<02:42, 58.48it/s]




 81%|████████  | 40506/50000 [11:02<02:40, 59.23it/s]




 81%|████████  | 40513/50000 [11:02<02:35, 61.16it/s]




 81%|████████  | 40520/50000 [11:03<02:38, 59.82it/s]




 81%|████████  | 40528/50000 [11:03<02:31, 62.39it/s]




 81%|████████  | 40535/50000 [11:03<02:27, 64.19it/s]




 81%|████████  | 40542/50000 [11:03<02:25, 64.78it/s]




 81%|████████  | 40550/50000 [11:03<02:21, 66.93it/s]




 81%|████████  | 40557/50000 [11:03<02:20, 67.18it/s]




 81%|████████  | 40565/50000 [11:03<02:18, 68.11it/s]




 81%|████████  | 40572/50000 [11:03<02:21, 66.66it/s]




 81%|████████  | 40579/50000 [11:03<02:24, 65.23it/s]




 81%|████████  | 40586/50000 [11:04<02:38, 59.52it/s]




 81%|████████  | 40593/50000 [11:04<02:31, 62.20it/s]




 81%|████████  | 40600/50000 [11:04<02:31, 61.84it/s]




 81%|████████  | 40607/50000 [11:04<02:29, 62.75it/s]




 81%|████████  | 40615/50000 [11:04<02:24, 64.86it/s]




 81%|████████  | 40622/50000 [11:04<02:25, 64.36it/s]




 81%|████████▏ | 40629/50000 [11:04<02:28, 62.98it/s]




 81%|████████▏ | 40636/50000 [11:04<02:51, 54.59it/s]




 81%|████████▏ | 40643/50000 [11:05<02:53, 53.78it/s]




 81%|████████▏ | 40649/50000 [11:05<02:49, 55.08it/s]




 81%|████████▏ | 40655/50000 [11:05<02:46, 56.29it/s]




 81%|████████▏ | 40661/50000 [11:05<02:45, 56.34it/s]




 81%|████████▏ | 40667/50000 [11:05<03:00, 51.67it/s]




 81%|████████▏ | 40673/50000 [11:05<02:53, 53.81it/s]




 81%|████████▏ | 40679/50000 [11:05<02:53, 53.64it/s]




 81%|████████▏ | 40685/50000 [11:05<02:53, 53.66it/s]




 81%|████████▏ | 40691/50000 [11:05<03:03, 50.69it/s]




 81%|████████▏ | 40697/50000 [11:06<03:09, 49.04it/s]




 81%|████████▏ | 40702/50000 [11:06<03:24, 45.47it/s]




 81%|████████▏ | 40708/50000 [11:06<03:15, 47.44it/s]




 81%|████████▏ | 40713/50000 [11:06<03:23, 45.71it/s]




 81%|████████▏ | 40720/50000 [11:06<03:02, 50.83it/s]




 81%|████████▏ | 40726/50000 [11:06<02:55, 52.85it/s]




 81%|████████▏ | 40732/50000 [11:06<02:49, 54.77it/s]




 81%|████████▏ | 40738/50000 [11:06<02:46, 55.58it/s]




 81%|████████▏ | 40745/50000 [11:06<02:41, 57.39it/s]




 82%|████████▏ | 40752/50000 [11:07<02:37, 58.88it/s]




 82%|████████▏ | 40759/50000 [11:07<02:36, 59.21it/s]




 82%|████████▏ | 40765/50000 [11:07<02:42, 56.79it/s]




 82%|████████▏ | 40773/50000 [11:07<02:32, 60.42it/s]




 82%|████████▏ | 40780/50000 [11:07<02:37, 58.63it/s]




 82%|████████▏ | 40786/50000 [11:07<02:40, 57.39it/s]




 82%|████████▏ | 40792/50000 [11:07<02:39, 57.86it/s]




 82%|████████▏ | 40799/50000 [11:07<02:36, 58.63it/s]




 82%|████████▏ | 40806/50000 [11:08<02:34, 59.33it/s]




 82%|████████▏ | 40813/50000 [11:08<02:28, 61.72it/s]




 82%|████████▏ | 40820/50000 [11:08<02:33, 59.86it/s]




 82%|████████▏ | 40827/50000 [11:08<02:43, 56.27it/s]




 82%|████████▏ | 40834/50000 [11:08<02:35, 59.00it/s]




 82%|████████▏ | 40840/50000 [11:08<02:35, 59.08it/s]




 82%|████████▏ | 40846/50000 [11:08<02:41, 56.55it/s]




 82%|████████▏ | 40853/50000 [11:08<02:37, 58.11it/s]




 82%|████████▏ | 40859/50000 [11:08<02:38, 57.53it/s]




 82%|████████▏ | 40865/50000 [11:09<02:48, 54.20it/s]




 82%|████████▏ | 40871/50000 [11:09<02:44, 55.40it/s]




 82%|████████▏ | 40877/50000 [11:09<02:51, 53.28it/s]




 82%|████████▏ | 40884/50000 [11:09<02:42, 56.05it/s]




 82%|████████▏ | 40890/50000 [11:09<02:46, 54.58it/s]




 82%|████████▏ | 40896/50000 [11:09<02:56, 51.66it/s]




 82%|████████▏ | 40902/50000 [11:09<03:14, 46.78it/s]




 82%|████████▏ | 40908/50000 [11:09<03:02, 49.88it/s]




 82%|████████▏ | 40914/50000 [11:10<03:12, 47.08it/s]




 82%|████████▏ | 40919/50000 [11:10<03:34, 42.36it/s]




 82%|████████▏ | 40925/50000 [11:10<03:21, 45.03it/s]




 82%|████████▏ | 40930/50000 [11:10<03:16, 46.06it/s]




 82%|████████▏ | 40936/50000 [11:10<03:11, 47.32it/s]




 82%|████████▏ | 40941/50000 [11:10<03:09, 47.86it/s]




 82%|████████▏ | 40946/50000 [11:10<03:09, 47.82it/s]




 82%|████████▏ | 40951/50000 [11:10<03:13, 46.73it/s]




 82%|████████▏ | 40956/50000 [11:10<03:22, 44.64it/s]




 82%|████████▏ | 40961/50000 [11:11<03:29, 43.12it/s]




 82%|████████▏ | 40968/50000 [11:11<03:11, 47.09it/s]




 82%|████████▏ | 40973/50000 [11:11<03:11, 47.08it/s]




 82%|████████▏ | 40979/50000 [11:11<03:05, 48.54it/s]




 82%|████████▏ | 40985/50000 [11:11<02:56, 51.00it/s]




 82%|████████▏ | 40991/50000 [11:11<02:53, 52.01it/s]




 82%|████████▏ | 40998/50000 [11:11<02:44, 54.87it/s]




 82%|████████▏ | 41005/50000 [11:11<02:37, 57.20it/s]




 82%|████████▏ | 41012/50000 [11:11<02:32, 59.04it/s]




 82%|████████▏ | 41019/50000 [11:12<02:32, 59.02it/s]




 82%|████████▏ | 41025/50000 [11:12<02:42, 55.12it/s]




 82%|████████▏ | 41031/50000 [11:12<02:51, 52.43it/s]




 82%|████████▏ | 41038/50000 [11:12<02:44, 54.49it/s]




 82%|████████▏ | 41044/50000 [11:12<02:54, 51.21it/s]




 82%|████████▏ | 41050/50000 [11:12<03:02, 49.01it/s]




 82%|████████▏ | 41055/50000 [11:12<03:18, 44.96it/s]




 82%|████████▏ | 41061/50000 [11:12<03:15, 45.76it/s]




 82%|████████▏ | 41066/50000 [11:13<03:18, 45.09it/s]




 82%|████████▏ | 41072/50000 [11:13<03:05, 48.17it/s]




 82%|████████▏ | 41079/50000 [11:13<02:52, 51.65it/s]




 82%|████████▏ | 41085/50000 [11:13<02:47, 53.08it/s]




 82%|████████▏ | 41092/50000 [11:13<02:39, 55.77it/s]




 82%|████████▏ | 41099/50000 [11:13<02:34, 57.66it/s]




 82%|████████▏ | 41106/50000 [11:13<02:33, 57.76it/s]




 82%|████████▏ | 41112/50000 [11:13<02:43, 54.27it/s]




 82%|████████▏ | 41118/50000 [11:14<02:51, 51.86it/s]




 82%|████████▏ | 41125/50000 [11:14<02:43, 54.31it/s]




 82%|████████▏ | 41132/50000 [11:14<02:34, 57.41it/s]




 82%|████████▏ | 41140/50000 [11:14<02:24, 61.46it/s]




 82%|████████▏ | 41147/50000 [11:14<02:22, 62.15it/s]




 82%|████████▏ | 41154/50000 [11:14<02:21, 62.47it/s]




 82%|████████▏ | 41161/50000 [11:14<02:20, 63.03it/s]




 82%|████████▏ | 41168/50000 [11:14<02:23, 61.68it/s]




 82%|████████▏ | 41175/50000 [11:14<02:27, 59.91it/s]




 82%|████████▏ | 41182/50000 [11:15<02:39, 55.45it/s]




 82%|████████▏ | 41188/50000 [11:15<02:44, 53.60it/s]




 82%|████████▏ | 41194/50000 [11:15<02:51, 51.30it/s]




 82%|████████▏ | 41201/50000 [11:15<02:42, 54.06it/s]




 82%|████████▏ | 41209/50000 [11:15<02:30, 58.48it/s]




 82%|████████▏ | 41216/50000 [11:15<02:27, 59.37it/s]




 82%|████████▏ | 41223/50000 [11:15<02:21, 61.92it/s]




 82%|████████▏ | 41230/50000 [11:15<02:26, 59.76it/s]




 82%|████████▏ | 41237/50000 [11:16<02:26, 59.98it/s]




 82%|████████▏ | 41244/50000 [11:16<02:24, 60.45it/s]




 83%|████████▎ | 41251/50000 [11:16<02:22, 61.42it/s]




 83%|████████▎ | 41258/50000 [11:16<02:17, 63.46it/s]




 83%|████████▎ | 41265/50000 [11:16<02:18, 63.22it/s]




 83%|████████▎ | 41273/50000 [11:16<02:12, 66.01it/s]




 83%|████████▎ | 41280/50000 [11:16<02:11, 66.07it/s]




 83%|████████▎ | 41287/50000 [11:16<02:18, 63.08it/s]




 83%|████████▎ | 41294/50000 [11:16<02:17, 63.38it/s]




 83%|████████▎ | 41301/50000 [11:16<02:16, 63.86it/s]




 83%|████████▎ | 41308/50000 [11:17<02:23, 60.69it/s]




 83%|████████▎ | 41315/50000 [11:17<02:24, 60.02it/s]




 83%|████████▎ | 41322/50000 [11:17<02:23, 60.47it/s]




 83%|████████▎ | 41329/50000 [11:17<02:24, 60.02it/s]




 83%|████████▎ | 41336/50000 [11:17<02:18, 62.41it/s]




 83%|████████▎ | 41343/50000 [11:17<02:18, 62.65it/s]




 83%|████████▎ | 41350/50000 [11:17<02:18, 62.66it/s]




 83%|████████▎ | 41357/50000 [11:17<02:22, 60.86it/s]




 83%|████████▎ | 41365/50000 [11:18<02:14, 64.35it/s]




 83%|████████▎ | 41372/50000 [11:18<02:20, 61.33it/s]




 83%|████████▎ | 41379/50000 [11:18<02:18, 62.12it/s]




 83%|████████▎ | 41386/50000 [11:18<02:16, 63.29it/s]




 83%|████████▎ | 41393/50000 [11:18<02:17, 62.51it/s]




 83%|████████▎ | 41400/50000 [11:18<02:13, 64.28it/s]




 83%|████████▎ | 41408/50000 [11:18<02:12, 64.71it/s]




 83%|████████▎ | 41415/50000 [11:18<02:16, 62.71it/s]




 83%|████████▎ | 41423/50000 [11:18<02:11, 65.46it/s]




 83%|████████▎ | 41431/50000 [11:19<02:05, 68.05it/s]




 83%|████████▎ | 41438/50000 [11:19<02:06, 67.68it/s]




 83%|████████▎ | 41446/50000 [11:19<02:03, 69.53it/s]




 83%|████████▎ | 41453/50000 [11:19<02:03, 69.11it/s]




 83%|████████▎ | 41460/50000 [11:19<02:07, 67.24it/s]




 83%|████████▎ | 41467/50000 [11:19<02:09, 65.98it/s]




 83%|████████▎ | 41474/50000 [11:19<02:09, 66.00it/s]




 83%|████████▎ | 41481/50000 [11:19<02:06, 67.09it/s]




 83%|████████▎ | 41488/50000 [11:19<02:06, 67.54it/s]




 83%|████████▎ | 41495/50000 [11:19<02:06, 67.23it/s]




 83%|████████▎ | 41502/50000 [11:20<02:21, 59.99it/s]




 83%|████████▎ | 41509/50000 [11:20<02:23, 59.31it/s]




 83%|████████▎ | 41516/50000 [11:20<02:18, 61.30it/s]




 83%|████████▎ | 41523/50000 [11:20<02:15, 62.36it/s]




 83%|████████▎ | 41530/50000 [11:20<02:16, 62.12it/s]




 83%|████████▎ | 41537/50000 [11:20<02:13, 63.47it/s]




 83%|████████▎ | 41544/50000 [11:20<02:24, 58.48it/s]




 83%|████████▎ | 41550/50000 [11:20<02:25, 58.12it/s]




 83%|████████▎ | 41557/50000 [11:21<02:19, 60.33it/s]




 83%|████████▎ | 41565/50000 [11:21<02:12, 63.78it/s]




 83%|████████▎ | 41572/50000 [11:21<02:09, 65.30it/s]




 83%|████████▎ | 41579/50000 [11:21<02:20, 60.02it/s]




 83%|████████▎ | 41586/50000 [11:21<02:26, 57.57it/s]




 83%|████████▎ | 41594/50000 [11:21<02:17, 61.03it/s]




 83%|████████▎ | 41601/50000 [11:21<02:12, 63.36it/s]




 83%|████████▎ | 41608/50000 [11:21<02:11, 64.01it/s]




 83%|████████▎ | 41616/50000 [11:21<02:06, 66.45it/s]




 83%|████████▎ | 41623/50000 [11:22<02:09, 64.90it/s]




 83%|████████▎ | 41630/50000 [11:22<02:13, 62.84it/s]




 83%|████████▎ | 41637/50000 [11:22<02:15, 61.63it/s]




 83%|████████▎ | 41644/50000 [11:22<02:31, 55.21it/s]




 83%|████████▎ | 41651/50000 [11:22<02:27, 56.43it/s]




 83%|████████▎ | 41659/50000 [11:22<02:18, 60.26it/s]




 83%|████████▎ | 41666/50000 [11:22<02:19, 59.95it/s]




 83%|████████▎ | 41674/50000 [11:22<02:11, 63.33it/s]




 83%|████████▎ | 41681/50000 [11:23<02:08, 64.52it/s]




 83%|████████▎ | 41688/50000 [11:23<02:11, 63.39it/s]




 83%|████████▎ | 41695/50000 [11:23<02:12, 62.58it/s]




 83%|████████▎ | 41702/50000 [11:23<02:13, 62.19it/s]




 83%|████████▎ | 41709/50000 [11:23<02:12, 62.41it/s]




 83%|████████▎ | 41716/50000 [11:23<02:15, 61.34it/s]




 83%|████████▎ | 41723/50000 [11:23<02:12, 62.56it/s]




 83%|████████▎ | 41730/50000 [11:23<02:12, 62.42it/s]




 83%|████████▎ | 41737/50000 [11:23<02:12, 62.16it/s]




 83%|████████▎ | 41744/50000 [11:24<02:17, 60.07it/s]




 84%|████████▎ | 41751/50000 [11:24<02:16, 60.36it/s]




 84%|████████▎ | 41758/50000 [11:24<02:24, 57.16it/s]




 84%|████████▎ | 41765/50000 [11:24<02:25, 56.64it/s]




 84%|████████▎ | 41771/50000 [11:24<02:28, 55.42it/s]




 84%|████████▎ | 41779/50000 [11:24<02:17, 59.59it/s]




 84%|████████▎ | 41787/50000 [11:24<02:10, 62.88it/s]




 84%|████████▎ | 41794/50000 [11:24<02:08, 63.98it/s]




 84%|████████▎ | 41801/50000 [11:24<02:05, 65.54it/s]




 84%|████████▎ | 41808/50000 [11:25<02:09, 63.16it/s]




 84%|████████▎ | 41815/50000 [11:25<02:08, 63.53it/s]




 84%|████████▎ | 41822/50000 [11:25<02:12, 61.93it/s]




 84%|████████▎ | 41830/50000 [11:25<02:03, 66.31it/s]




 84%|████████▎ | 41837/50000 [11:25<02:11, 61.90it/s]




 84%|████████▎ | 41844/50000 [11:25<02:23, 56.98it/s]




 84%|████████▎ | 41851/50000 [11:25<02:24, 56.46it/s]




 84%|████████▎ | 41858/50000 [11:25<02:23, 56.91it/s]




 84%|████████▎ | 41864/50000 [11:26<02:37, 51.73it/s]




 84%|████████▎ | 41870/50000 [11:26<02:34, 52.72it/s]




 84%|████████▍ | 41876/50000 [11:26<02:35, 52.29it/s]




 84%|████████▍ | 41882/50000 [11:26<02:35, 52.16it/s]




 84%|████████▍ | 41888/50000 [11:26<02:44, 49.26it/s]




 84%|████████▍ | 41894/50000 [11:26<02:39, 50.77it/s]




 84%|████████▍ | 41900/50000 [11:26<02:39, 50.83it/s]




 84%|████████▍ | 41906/50000 [11:26<02:35, 52.07it/s]




 84%|████████▍ | 41912/50000 [11:27<02:31, 53.53it/s]




 84%|████████▍ | 41919/50000 [11:27<02:27, 54.93it/s]




 84%|████████▍ | 41926/50000 [11:27<02:19, 57.89it/s]




 84%|████████▍ | 41933/50000 [11:27<02:20, 57.57it/s]




 84%|████████▍ | 41940/50000 [11:27<02:17, 58.42it/s]




 84%|████████▍ | 41947/50000 [11:27<02:13, 60.45it/s]




 84%|████████▍ | 41954/50000 [11:27<02:15, 59.39it/s]




 84%|████████▍ | 41960/50000 [11:27<02:16, 58.75it/s]




 84%|████████▍ | 41966/50000 [11:27<02:28, 54.05it/s]




 84%|████████▍ | 41972/50000 [11:28<02:28, 54.02it/s]




 84%|████████▍ | 41978/50000 [11:28<02:33, 52.30it/s]




 84%|████████▍ | 41984/50000 [11:28<02:29, 53.47it/s]




 84%|████████▍ | 41990/50000 [11:28<02:30, 53.33it/s]




 84%|████████▍ | 41996/50000 [11:28<02:36, 51.26it/s]




 84%|████████▍ | 42002/50000 [11:28<02:42, 49.29it/s]




 84%|████████▍ | 42007/50000 [11:28<02:44, 48.53it/s]




 84%|████████▍ | 42014/50000 [11:28<02:32, 52.29it/s]




 84%|████████▍ | 42020/50000 [11:28<02:30, 53.13it/s]




 84%|████████▍ | 42027/50000 [11:29<02:20, 56.68it/s]




 84%|████████▍ | 42033/50000 [11:29<02:19, 57.10it/s]




 84%|████████▍ | 42040/50000 [11:29<02:14, 59.26it/s]




 84%|████████▍ | 42047/50000 [11:29<02:16, 58.15it/s]




 84%|████████▍ | 42055/50000 [11:29<02:05, 63.08it/s]




 84%|████████▍ | 42062/50000 [11:29<02:06, 62.86it/s]




 84%|████████▍ | 42069/50000 [11:29<02:08, 61.85it/s]




 84%|████████▍ | 42076/50000 [11:29<02:04, 63.61it/s]




 84%|████████▍ | 42083/50000 [11:29<02:07, 61.98it/s]




 84%|████████▍ | 42090/50000 [11:30<02:05, 63.02it/s]




 84%|████████▍ | 42097/50000 [11:30<02:07, 61.75it/s]




 84%|████████▍ | 42105/50000 [11:30<02:01, 64.88it/s]




 84%|████████▍ | 42113/50000 [11:30<01:59, 65.78it/s]




 84%|████████▍ | 42120/50000 [11:30<02:01, 64.64it/s]




 84%|████████▍ | 42128/50000 [11:30<01:56, 67.77it/s]




 84%|████████▍ | 42135/50000 [11:30<02:00, 65.23it/s]




 84%|████████▍ | 42142/50000 [11:30<01:58, 66.36it/s]




 84%|████████▍ | 42149/50000 [11:30<01:58, 66.51it/s]




 84%|████████▍ | 42156/50000 [11:31<01:59, 65.49it/s]




 84%|████████▍ | 42163/50000 [11:31<01:58, 66.08it/s]




 84%|████████▍ | 42170/50000 [11:31<02:05, 62.50it/s]




 84%|████████▍ | 42177/50000 [11:31<02:05, 62.13it/s]




 84%|████████▍ | 42184/50000 [11:31<02:07, 61.47it/s]




 84%|████████▍ | 42191/50000 [11:31<02:04, 62.49it/s]




 84%|████████▍ | 42198/50000 [11:31<02:05, 62.37it/s]




 84%|████████▍ | 42205/50000 [11:31<02:02, 63.82it/s]




 84%|████████▍ | 42212/50000 [11:31<02:09, 60.36it/s]




 84%|████████▍ | 42219/50000 [11:32<02:16, 57.16it/s]




 84%|████████▍ | 42225/50000 [11:32<02:16, 57.04it/s]




 84%|████████▍ | 42232/50000 [11:32<02:13, 58.03it/s]




 84%|████████▍ | 42238/50000 [11:32<02:20, 55.41it/s]




 84%|████████▍ | 42245/50000 [11:32<02:12, 58.56it/s]




 85%|████████▍ | 42252/50000 [11:32<02:06, 61.46it/s]




 85%|████████▍ | 42259/50000 [11:32<02:07, 60.77it/s]




 85%|████████▍ | 42266/50000 [11:32<02:09, 59.61it/s]




 85%|████████▍ | 42273/50000 [11:33<02:08, 60.18it/s]




 85%|████████▍ | 42280/50000 [11:33<02:08, 60.28it/s]




 85%|████████▍ | 42287/50000 [11:33<02:04, 62.11it/s]




 85%|████████▍ | 42294/50000 [11:33<02:00, 64.15it/s]




 85%|████████▍ | 42301/50000 [11:33<02:01, 63.36it/s]




 85%|████████▍ | 42308/50000 [11:33<02:01, 63.32it/s]




 85%|████████▍ | 42315/50000 [11:33<02:03, 62.12it/s]




 85%|████████▍ | 42322/50000 [11:33<02:04, 61.79it/s]




 85%|████████▍ | 42329/50000 [11:33<02:01, 63.22it/s]




 85%|████████▍ | 42336/50000 [11:34<02:05, 61.08it/s]




 85%|████████▍ | 42343/50000 [11:34<02:01, 62.97it/s]




 85%|████████▍ | 42351/50000 [11:34<01:55, 66.10it/s]




 85%|████████▍ | 42358/50000 [11:34<01:57, 65.12it/s]




 85%|████████▍ | 42365/50000 [11:34<01:59, 64.10it/s]




 85%|████████▍ | 42372/50000 [11:34<02:01, 62.81it/s]




 85%|████████▍ | 42379/50000 [11:34<02:08, 59.27it/s]




 85%|████████▍ | 42385/50000 [11:34<02:09, 58.83it/s]




 85%|████████▍ | 42391/50000 [11:34<02:18, 55.00it/s]




 85%|████████▍ | 42397/50000 [11:35<02:22, 53.44it/s]




 85%|████████▍ | 42403/50000 [11:35<02:17, 55.14it/s]




 85%|████████▍ | 42409/50000 [11:35<02:16, 55.77it/s]




 85%|████████▍ | 42416/50000 [11:35<02:10, 58.17it/s]




 85%|████████▍ | 42422/50000 [11:35<02:12, 57.15it/s]




 85%|████████▍ | 42429/50000 [11:35<02:07, 59.60it/s]




 85%|████████▍ | 42436/50000 [11:35<02:15, 55.73it/s]




 85%|████████▍ | 42442/50000 [11:35<02:15, 55.72it/s]




 85%|████████▍ | 42448/50000 [11:35<02:14, 56.26it/s]




 85%|████████▍ | 42454/50000 [11:36<02:19, 54.27it/s]




 85%|████████▍ | 42460/50000 [11:36<02:15, 55.76it/s]




 85%|████████▍ | 42467/50000 [11:36<02:07, 59.27it/s]




 85%|████████▍ | 42474/50000 [11:36<02:13, 56.33it/s]




 85%|████████▍ | 42480/50000 [11:36<02:18, 54.47it/s]




 85%|████████▍ | 42486/50000 [11:36<02:14, 55.75it/s]




 85%|████████▍ | 42492/50000 [11:36<02:30, 49.91it/s]




 85%|████████▍ | 42499/50000 [11:36<02:19, 53.77it/s]




 85%|████████▌ | 42506/50000 [11:37<02:12, 56.43it/s]




 85%|████████▌ | 42512/50000 [11:37<02:13, 56.05it/s]




 85%|████████▌ | 42519/50000 [11:37<02:06, 59.25it/s]




 85%|████████▌ | 42527/50000 [11:37<01:58, 62.93it/s]




 85%|████████▌ | 42534/50000 [11:37<01:59, 62.50it/s]




 85%|████████▌ | 42542/50000 [11:37<01:55, 64.60it/s]




 85%|████████▌ | 42549/50000 [11:37<01:53, 65.90it/s]




 85%|████████▌ | 42556/50000 [11:37<01:55, 64.62it/s]




 85%|████████▌ | 42563/50000 [11:37<01:57, 63.16it/s]




 85%|████████▌ | 42571/50000 [11:37<01:54, 65.16it/s]




 85%|████████▌ | 42578/50000 [11:38<01:51, 66.40it/s]




 85%|████████▌ | 42585/50000 [11:38<01:51, 66.53it/s]




 85%|████████▌ | 42592/50000 [11:38<01:58, 62.53it/s]




 85%|████████▌ | 42599/50000 [11:38<01:59, 61.75it/s]




 85%|████████▌ | 42606/50000 [11:38<02:01, 61.05it/s]




 85%|████████▌ | 42613/50000 [11:38<01:57, 63.10it/s]




 85%|████████▌ | 42620/50000 [11:38<01:55, 63.65it/s]




 85%|████████▌ | 42627/50000 [11:38<01:57, 62.84it/s]




 85%|████████▌ | 42634/50000 [11:39<01:59, 61.63it/s]




 85%|████████▌ | 42641/50000 [11:39<02:04, 58.97it/s]




 85%|████████▌ | 42649/50000 [11:39<01:56, 62.93it/s]




 85%|████████▌ | 42656/50000 [11:39<01:54, 63.96it/s]




 85%|████████▌ | 42663/50000 [11:39<02:03, 59.38it/s]




 85%|████████▌ | 42670/50000 [11:39<02:01, 60.48it/s]




 85%|████████▌ | 42677/50000 [11:39<01:57, 62.43it/s]




 85%|████████▌ | 42684/50000 [11:39<01:59, 61.35it/s]




 85%|████████▌ | 42691/50000 [11:39<01:58, 61.90it/s]




 85%|████████▌ | 42698/50000 [11:40<02:02, 59.75it/s]




 85%|████████▌ | 42705/50000 [11:40<02:03, 59.22it/s]




 85%|████████▌ | 42713/50000 [11:40<01:54, 63.80it/s]




 85%|████████▌ | 42720/50000 [11:40<01:54, 63.46it/s]




 85%|████████▌ | 42727/50000 [11:40<01:56, 62.21it/s]




 85%|████████▌ | 42734/50000 [11:40<01:54, 63.24it/s]




 85%|████████▌ | 42741/50000 [11:40<01:53, 64.19it/s]




 85%|████████▌ | 42749/50000 [11:40<01:47, 67.17it/s]




 86%|████████▌ | 42756/50000 [11:40<01:52, 64.39it/s]




 86%|████████▌ | 42764/50000 [11:41<01:47, 67.58it/s]




 86%|████████▌ | 42771/50000 [11:41<01:48, 66.59it/s]




 86%|████████▌ | 42778/50000 [11:41<01:54, 63.07it/s]




 86%|████████▌ | 42785/50000 [11:41<02:09, 55.62it/s]




 86%|████████▌ | 42791/50000 [11:41<02:25, 49.71it/s]




 86%|████████▌ | 42797/50000 [11:41<02:24, 49.96it/s]




 86%|████████▌ | 42803/50000 [11:41<02:26, 48.97it/s]




 86%|████████▌ | 42809/50000 [11:41<02:34, 46.48it/s]




 86%|████████▌ | 42814/50000 [11:42<02:34, 46.59it/s]




 86%|████████▌ | 42819/50000 [11:42<02:33, 46.80it/s]




 86%|████████▌ | 42824/50000 [11:42<02:46, 43.14it/s]




 86%|████████▌ | 42829/50000 [11:42<02:44, 43.67it/s]




 86%|████████▌ | 42835/50000 [11:42<02:35, 46.02it/s]




 86%|████████▌ | 42840/50000 [11:42<02:37, 45.57it/s]




 86%|████████▌ | 42846/50000 [11:42<02:30, 47.63it/s]




 86%|████████▌ | 42851/50000 [11:42<02:29, 47.80it/s]




 86%|████████▌ | 42856/50000 [11:42<02:28, 48.20it/s]




 86%|████████▌ | 42861/50000 [11:43<02:28, 47.92it/s]




 86%|████████▌ | 42868/50000 [11:43<02:19, 51.11it/s]




 86%|████████▌ | 42876/50000 [11:43<02:10, 54.55it/s]




 86%|████████▌ | 42882/50000 [11:43<02:08, 55.50it/s]




 86%|████████▌ | 42890/50000 [11:43<01:57, 60.39it/s]




 86%|████████▌ | 42897/50000 [11:43<01:56, 61.05it/s]




 86%|████████▌ | 42904/50000 [11:43<01:58, 59.95it/s]




 86%|████████▌ | 42911/50000 [11:43<01:58, 59.81it/s]




 86%|████████▌ | 42918/50000 [11:44<01:56, 60.80it/s]




 86%|████████▌ | 42925/50000 [11:44<02:18, 50.99it/s]




 86%|████████▌ | 42932/50000 [11:44<02:09, 54.45it/s]




 86%|████████▌ | 42938/50000 [11:44<02:14, 52.66it/s]




 86%|████████▌ | 42944/50000 [11:44<02:25, 48.37it/s]




 86%|████████▌ | 42950/50000 [11:44<02:33, 45.97it/s]




 86%|████████▌ | 42956/50000 [11:44<02:25, 48.40it/s]




 86%|████████▌ | 42961/50000 [11:44<02:29, 46.98it/s]




 86%|████████▌ | 42966/50000 [11:45<02:28, 47.21it/s]




 86%|████████▌ | 42973/50000 [11:45<02:16, 51.43it/s]




 86%|████████▌ | 42980/50000 [11:45<02:07, 55.25it/s]




 86%|████████▌ | 42986/50000 [11:45<02:05, 55.85it/s]




 86%|████████▌ | 42992/50000 [11:45<02:06, 55.42it/s]




 86%|████████▌ | 42998/50000 [11:45<02:06, 55.50it/s]




 86%|████████▌ | 43006/50000 [11:45<01:55, 60.46it/s]




 86%|████████▌ | 43013/50000 [11:45<01:51, 62.75it/s]




 86%|████████▌ | 43020/50000 [11:45<01:48, 64.10it/s]




 86%|████████▌ | 43027/50000 [11:46<01:47, 64.90it/s]




 86%|████████▌ | 43034/50000 [11:46<01:47, 64.92it/s]




 86%|████████▌ | 43042/50000 [11:46<01:42, 67.81it/s]




 86%|████████▌ | 43049/50000 [11:46<01:43, 67.13it/s]




 86%|████████▌ | 43056/50000 [11:46<01:49, 63.24it/s]




 86%|████████▌ | 43063/50000 [11:46<01:52, 61.41it/s]




 86%|████████▌ | 43070/50000 [11:46<01:57, 58.98it/s]




 86%|████████▌ | 43076/50000 [11:46<02:09, 53.40it/s]




 86%|████████▌ | 43082/50000 [11:46<02:07, 54.14it/s]




 86%|████████▌ | 43088/50000 [11:47<02:06, 54.60it/s]




 86%|████████▌ | 43095/50000 [11:47<01:58, 58.35it/s]




 86%|████████▌ | 43104/50000 [11:47<01:47, 64.16it/s]




 86%|████████▌ | 43111/50000 [11:47<01:46, 64.69it/s]




 86%|████████▌ | 43120/50000 [11:47<01:39, 69.22it/s]




 86%|████████▋ | 43128/50000 [11:47<01:49, 63.00it/s]




 86%|████████▋ | 43135/50000 [11:47<01:46, 64.72it/s]




 86%|████████▋ | 43142/50000 [11:47<01:49, 62.55it/s]




 86%|████████▋ | 43149/50000 [11:47<01:55, 59.41it/s]




 86%|████████▋ | 43156/50000 [11:48<02:11, 52.03it/s]




 86%|████████▋ | 43163/50000 [11:48<02:02, 55.73it/s]




 86%|████████▋ | 43169/50000 [11:48<02:01, 56.19it/s]




 86%|████████▋ | 43176/50000 [11:48<01:55, 58.87it/s]




 86%|████████▋ | 43183/50000 [11:48<01:54, 59.50it/s]




 86%|████████▋ | 43190/50000 [11:48<01:52, 60.65it/s]




 86%|████████▋ | 43197/50000 [11:48<01:58, 57.62it/s]




 86%|████████▋ | 43204/50000 [11:48<01:55, 59.05it/s]




 86%|████████▋ | 43211/50000 [11:49<01:52, 60.16it/s]




 86%|████████▋ | 43218/50000 [11:49<01:48, 62.47it/s]




 86%|████████▋ | 43226/50000 [11:49<01:45, 64.41it/s]




 86%|████████▋ | 43233/50000 [11:49<01:52, 60.33it/s]




 86%|████████▋ | 43240/50000 [11:49<01:54, 58.87it/s]




 86%|████████▋ | 43246/50000 [11:49<01:54, 58.90it/s]




 87%|████████▋ | 43253/50000 [11:49<01:50, 60.92it/s]




 87%|████████▋ | 43261/50000 [11:49<01:46, 63.32it/s]




 87%|████████▋ | 43268/50000 [11:49<01:51, 60.36it/s]




 87%|████████▋ | 43275/50000 [11:50<01:56, 57.72it/s]




 87%|████████▋ | 43284/50000 [11:50<01:46, 63.22it/s]




 87%|████████▋ | 43291/50000 [11:50<01:43, 64.80it/s]




 87%|████████▋ | 43299/50000 [11:50<01:39, 67.21it/s]




 87%|████████▋ | 43306/50000 [11:50<01:45, 63.29it/s]




 87%|████████▋ | 43313/50000 [11:50<01:49, 61.28it/s]




 87%|████████▋ | 43320/50000 [11:50<01:46, 62.93it/s]




 87%|████████▋ | 43327/50000 [11:50<01:51, 60.11it/s]




 87%|████████▋ | 43334/50000 [11:51<01:50, 60.38it/s]




 87%|████████▋ | 43342/50000 [11:51<01:44, 63.67it/s]




 87%|████████▋ | 43349/50000 [11:51<01:48, 61.05it/s]




 87%|████████▋ | 43356/50000 [11:51<01:47, 61.53it/s]




 87%|████████▋ | 43363/50000 [11:51<01:44, 63.72it/s]




 87%|████████▋ | 43370/50000 [11:51<01:41, 65.16it/s]




 87%|████████▋ | 43377/50000 [11:51<01:48, 60.95it/s]




 87%|████████▋ | 43384/50000 [11:51<01:54, 57.74it/s]




 87%|████████▋ | 43391/50000 [11:51<01:50, 59.89it/s]




 87%|████████▋ | 43398/50000 [11:52<01:52, 58.72it/s]




 87%|████████▋ | 43405/50000 [11:52<01:49, 59.98it/s]




 87%|████████▋ | 43412/50000 [11:52<01:48, 60.68it/s]




 87%|████████▋ | 43419/50000 [11:52<01:44, 62.68it/s]




 87%|████████▋ | 43426/50000 [11:52<01:44, 62.92it/s]




 87%|████████▋ | 43433/50000 [11:52<01:50, 59.56it/s]




 87%|████████▋ | 43440/50000 [11:52<01:47, 61.09it/s]




 87%|████████▋ | 43448/50000 [11:52<01:43, 63.46it/s]




 87%|████████▋ | 43455/50000 [11:52<01:41, 64.61it/s]




 87%|████████▋ | 43462/50000 [11:53<01:41, 64.55it/s]




 87%|████████▋ | 43469/50000 [11:53<01:40, 65.04it/s]




 87%|████████▋ | 43476/50000 [11:53<01:45, 62.09it/s]




 87%|████████▋ | 43483/50000 [11:53<01:43, 62.93it/s]




 87%|████████▋ | 43490/50000 [11:53<01:42, 63.45it/s]




 87%|████████▋ | 43497/50000 [11:53<01:50, 58.99it/s]




 87%|████████▋ | 43503/50000 [11:53<01:52, 57.63it/s]




 87%|████████▋ | 43510/50000 [11:53<01:47, 60.12it/s]




 87%|████████▋ | 43517/50000 [11:53<01:44, 61.78it/s]




 87%|████████▋ | 43524/50000 [11:54<01:47, 60.05it/s]




 87%|████████▋ | 43532/50000 [11:54<01:40, 64.55it/s]




 87%|████████▋ | 43539/50000 [11:54<01:42, 62.94it/s]




 87%|████████▋ | 43546/50000 [11:54<01:40, 64.42it/s]




 87%|████████▋ | 43553/50000 [11:54<01:43, 62.52it/s]




 87%|████████▋ | 43561/50000 [11:54<01:39, 64.84it/s]




 87%|████████▋ | 43568/50000 [11:54<01:39, 64.35it/s]




 87%|████████▋ | 43575/50000 [11:54<01:38, 65.26it/s]




 87%|████████▋ | 43584/50000 [11:54<01:32, 69.36it/s]




 87%|████████▋ | 43592/50000 [11:55<01:35, 67.12it/s]




 87%|████████▋ | 43599/50000 [11:55<01:41, 62.89it/s]




 87%|████████▋ | 43606/50000 [11:55<01:40, 63.68it/s]




 87%|████████▋ | 43613/50000 [11:55<01:40, 63.72it/s]




 87%|████████▋ | 43620/50000 [11:55<01:38, 64.70it/s]




 87%|████████▋ | 43627/50000 [11:55<01:44, 60.82it/s]




 87%|████████▋ | 43634/50000 [11:55<01:45, 60.10it/s]




 87%|████████▋ | 43641/50000 [11:55<01:53, 55.91it/s]




 87%|████████▋ | 43647/50000 [11:56<01:55, 55.23it/s]




 87%|████████▋ | 43654/50000 [11:56<01:49, 57.98it/s]




 87%|████████▋ | 43663/50000 [11:56<01:40, 63.17it/s]




 87%|████████▋ | 43670/50000 [11:56<01:43, 60.89it/s]




 87%|████████▋ | 43677/50000 [11:56<01:46, 59.38it/s]




 87%|████████▋ | 43684/50000 [11:56<01:56, 54.18it/s]




 87%|████████▋ | 43691/50000 [11:56<01:49, 57.59it/s]




 87%|████████▋ | 43697/50000 [11:56<01:48, 57.93it/s]




 87%|████████▋ | 43705/50000 [11:57<01:41, 61.88it/s]




 87%|████████▋ | 43712/50000 [11:57<01:46, 59.28it/s]




 87%|████████▋ | 43719/50000 [11:57<01:43, 60.73it/s]




 87%|████████▋ | 43726/50000 [11:57<01:44, 60.27it/s]




 87%|████████▋ | 43733/50000 [11:57<01:57, 53.20it/s]




 87%|████████▋ | 43739/50000 [11:57<02:00, 51.84it/s]




 87%|████████▋ | 43745/50000 [11:57<02:02, 51.05it/s]




 88%|████████▊ | 43751/50000 [11:57<02:03, 50.52it/s]




 88%|████████▊ | 43757/50000 [11:58<02:08, 48.45it/s]




 88%|████████▊ | 43763/50000 [11:58<02:11, 47.55it/s]




 88%|████████▊ | 43769/50000 [11:58<02:08, 48.53it/s]




 88%|████████▊ | 43774/50000 [11:58<02:08, 48.29it/s]




 88%|████████▊ | 43779/50000 [11:58<02:10, 47.85it/s]




 88%|████████▊ | 43784/50000 [11:58<02:22, 43.69it/s]




 88%|████████▊ | 43789/50000 [11:58<02:23, 43.38it/s]




 88%|████████▊ | 43794/50000 [11:58<02:19, 44.42it/s]




 88%|████████▊ | 43799/50000 [11:58<02:26, 42.43it/s]




 88%|████████▊ | 43804/50000 [11:59<02:22, 43.61it/s]




 88%|████████▊ | 43811/50000 [11:59<02:08, 48.09it/s]




 88%|████████▊ | 43817/50000 [11:59<02:05, 49.16it/s]




 88%|████████▊ | 43823/50000 [11:59<02:02, 50.57it/s]




 88%|████████▊ | 43830/50000 [11:59<01:55, 53.43it/s]




 88%|████████▊ | 43837/50000 [11:59<01:48, 56.86it/s]




 88%|████████▊ | 43844/50000 [11:59<01:44, 58.63it/s]




 88%|████████▊ | 43852/50000 [11:59<01:39, 62.00it/s]




 88%|████████▊ | 43859/50000 [11:59<01:38, 62.37it/s]




 88%|████████▊ | 43866/50000 [12:00<01:48, 56.57it/s]




 88%|████████▊ | 43872/50000 [12:00<01:55, 52.86it/s]




 88%|████████▊ | 43878/50000 [12:00<01:58, 51.77it/s]




 88%|████████▊ | 43884/50000 [12:00<01:56, 52.68it/s]




 88%|████████▊ | 43890/50000 [12:00<01:57, 52.02it/s]




 88%|████████▊ | 43896/50000 [12:00<01:59, 51.18it/s]




 88%|████████▊ | 43902/50000 [12:00<02:07, 47.83it/s]




 88%|████████▊ | 43907/50000 [12:01<02:11, 46.47it/s]




 88%|████████▊ | 43914/50000 [12:01<02:01, 50.16it/s]




 88%|████████▊ | 43920/50000 [12:01<01:56, 52.38it/s]




 88%|████████▊ | 43927/50000 [12:01<01:48, 56.15it/s]




 88%|████████▊ | 43933/50000 [12:01<01:50, 54.94it/s]




 88%|████████▊ | 43940/50000 [12:01<01:44, 57.97it/s]




 88%|████████▊ | 43947/50000 [12:01<01:42, 59.15it/s]




 88%|████████▊ | 43954/50000 [12:01<01:38, 61.67it/s]




 88%|████████▊ | 43961/50000 [12:01<01:39, 61.00it/s]




 88%|████████▊ | 43968/50000 [12:01<01:39, 60.38it/s]




 88%|████████▊ | 43975/50000 [12:02<01:38, 61.28it/s]




 88%|████████▊ | 43983/50000 [12:02<01:32, 64.82it/s]




 88%|████████▊ | 43990/50000 [12:02<01:33, 64.42it/s]




 88%|████████▊ | 43997/50000 [12:02<01:31, 65.86it/s]




 88%|████████▊ | 44004/50000 [12:02<01:34, 63.46it/s]




 88%|████████▊ | 44011/50000 [12:02<01:35, 62.71it/s]




 88%|████████▊ | 44018/50000 [12:02<01:37, 61.38it/s]




 88%|████████▊ | 44025/50000 [12:02<01:34, 62.93it/s]




 88%|████████▊ | 44032/50000 [12:02<01:33, 63.88it/s]




 88%|████████▊ | 44039/50000 [12:03<01:35, 62.66it/s]




 88%|████████▊ | 44046/50000 [12:03<01:35, 62.16it/s]




 88%|████████▊ | 44053/50000 [12:03<01:36, 61.46it/s]




 88%|████████▊ | 44060/50000 [12:03<01:35, 62.47it/s]




 88%|████████▊ | 44068/50000 [12:03<01:29, 66.25it/s]




 88%|████████▊ | 44075/50000 [12:03<01:31, 65.03it/s]




 88%|████████▊ | 44082/50000 [12:03<01:34, 62.49it/s]




 88%|████████▊ | 44090/50000 [12:03<01:29, 66.21it/s]




 88%|████████▊ | 44097/50000 [12:03<01:29, 66.11it/s]




 88%|████████▊ | 44104/50000 [12:04<01:31, 64.47it/s]




 88%|████████▊ | 44112/50000 [12:04<01:27, 66.91it/s]




 88%|████████▊ | 44120/50000 [12:04<01:27, 67.48it/s]




 88%|████████▊ | 44127/50000 [12:04<01:28, 66.52it/s]




 88%|████████▊ | 44134/50000 [12:04<01:28, 66.24it/s]




 88%|████████▊ | 44142/50000 [12:04<01:26, 67.60it/s]




 88%|████████▊ | 44150/50000 [12:04<01:24, 69.11it/s]




 88%|████████▊ | 44157/50000 [12:04<01:31, 64.10it/s]




 88%|████████▊ | 44164/50000 [12:05<01:32, 62.98it/s]




 88%|████████▊ | 44172/50000 [12:05<01:29, 65.19it/s]




 88%|████████▊ | 44179/50000 [12:05<01:34, 61.45it/s]




 88%|████████▊ | 44187/50000 [12:05<01:30, 63.93it/s]




 88%|████████▊ | 44194/50000 [12:05<01:33, 61.87it/s]




 88%|████████▊ | 44201/50000 [12:05<01:35, 60.50it/s]




 88%|████████▊ | 44208/50000 [12:05<01:34, 61.29it/s]




 88%|████████▊ | 44215/50000 [12:05<01:35, 60.26it/s]




 88%|████████▊ | 44223/50000 [12:05<01:31, 63.20it/s]




 88%|████████▊ | 44230/50000 [12:06<01:34, 61.06it/s]




 88%|████████▊ | 44238/50000 [12:06<01:29, 64.12it/s]




 88%|████████▊ | 44245/50000 [12:06<01:29, 64.64it/s]




 89%|████████▊ | 44253/50000 [12:06<01:25, 67.42it/s]




 89%|████████▊ | 44260/50000 [12:06<01:25, 66.87it/s]




 89%|████████▊ | 44267/50000 [12:06<01:26, 66.10it/s]




 89%|████████▊ | 44275/50000 [12:06<01:23, 68.71it/s]




 89%|████████▊ | 44282/50000 [12:06<01:24, 67.94it/s]




 89%|████████▊ | 44289/50000 [12:06<01:24, 67.22it/s]




 89%|████████▊ | 44296/50000 [12:07<01:26, 65.97it/s]




 89%|████████▊ | 44303/50000 [12:07<01:27, 64.94it/s]




 89%|████████▊ | 44310/50000 [12:07<01:25, 66.31it/s]




 89%|████████▊ | 44317/50000 [12:07<01:30, 62.90it/s]




 89%|████████▊ | 44324/50000 [12:07<01:30, 62.83it/s]




 89%|████████▊ | 44331/50000 [12:07<01:27, 64.69it/s]




 89%|████████▊ | 44338/50000 [12:07<01:28, 64.15it/s]




 89%|████████▊ | 44345/50000 [12:07<01:28, 64.23it/s]




 89%|████████▊ | 44352/50000 [12:07<01:29, 62.89it/s]




 89%|████████▊ | 44360/50000 [12:08<01:25, 65.76it/s]




 89%|████████▊ | 44367/50000 [12:08<01:26, 64.98it/s]




 89%|████████▊ | 44374/50000 [12:08<01:25, 65.71it/s]




 89%|████████▉ | 44382/50000 [12:08<01:21, 69.30it/s]




 89%|████████▉ | 44390/50000 [12:08<01:24, 66.17it/s]




 89%|████████▉ | 44398/50000 [12:08<01:21, 68.94it/s]




 89%|████████▉ | 44405/50000 [12:08<01:24, 66.17it/s]




 89%|████████▉ | 44412/50000 [12:08<01:23, 66.99it/s]




 89%|████████▉ | 44419/50000 [12:08<01:28, 63.32it/s]




 89%|████████▉ | 44427/50000 [12:09<01:25, 65.13it/s]




 89%|████████▉ | 44434/50000 [12:09<01:25, 65.23it/s]




 89%|████████▉ | 44441/50000 [12:09<01:27, 63.48it/s]




 89%|████████▉ | 44448/50000 [12:09<01:30, 61.65it/s]




 89%|████████▉ | 44455/50000 [12:09<01:38, 56.35it/s]




 89%|████████▉ | 44461/50000 [12:09<01:43, 53.60it/s]




 89%|████████▉ | 44468/50000 [12:09<01:37, 56.71it/s]




 89%|████████▉ | 44475/50000 [12:09<01:32, 59.72it/s]




 89%|████████▉ | 44482/50000 [12:09<01:30, 60.89it/s]




 89%|████████▉ | 44489/50000 [12:10<01:28, 62.23it/s]




 89%|████████▉ | 44496/50000 [12:10<01:27, 62.69it/s]




 89%|████████▉ | 44504/50000 [12:10<01:24, 65.29it/s]




 89%|████████▉ | 44511/50000 [12:10<01:23, 65.84it/s]




 89%|████████▉ | 44518/50000 [12:10<01:24, 65.03it/s]




 89%|████████▉ | 44525/50000 [12:10<01:23, 65.94it/s]




 89%|████████▉ | 44533/50000 [12:10<01:20, 67.89it/s]




 89%|████████▉ | 44540/50000 [12:10<01:26, 62.76it/s]




 89%|████████▉ | 44548/50000 [12:10<01:22, 65.74it/s]




 89%|████████▉ | 44555/50000 [12:11<01:28, 61.63it/s]




 89%|████████▉ | 44562/50000 [12:11<01:31, 59.19it/s]




 89%|████████▉ | 44569/50000 [12:11<01:30, 60.04it/s]




 89%|████████▉ | 44576/50000 [12:11<01:32, 58.67it/s]




 89%|████████▉ | 44582/50000 [12:11<01:36, 56.29it/s]




 89%|████████▉ | 44588/50000 [12:11<01:51, 48.63it/s]




 89%|████████▉ | 44596/50000 [12:11<01:39, 54.58it/s]




 89%|████████▉ | 44602/50000 [12:11<01:36, 55.99it/s]




 89%|████████▉ | 44609/50000 [12:12<01:33, 57.62it/s]




 89%|████████▉ | 44616/50000 [12:12<01:30, 59.65it/s]




 89%|████████▉ | 44623/50000 [12:12<01:31, 58.85it/s]




 89%|████████▉ | 44631/50000 [12:12<01:28, 60.62it/s]




 89%|████████▉ | 44639/50000 [12:12<01:24, 63.54it/s]




 89%|████████▉ | 44646/50000 [12:12<01:25, 62.68it/s]




 89%|████████▉ | 44653/50000 [12:12<01:36, 55.14it/s]




 89%|████████▉ | 44659/50000 [12:12<01:34, 56.24it/s]




 89%|████████▉ | 44665/50000 [12:13<01:33, 56.79it/s]




 89%|████████▉ | 44672/50000 [12:13<01:32, 57.57it/s]




 89%|████████▉ | 44679/50000 [12:13<01:29, 59.76it/s]




 89%|████████▉ | 44686/50000 [12:13<01:29, 59.68it/s]




 89%|████████▉ | 44693/50000 [12:13<01:34, 55.91it/s]




 89%|████████▉ | 44699/50000 [12:13<01:40, 52.63it/s]




 89%|████████▉ | 44705/50000 [12:13<01:50, 48.01it/s]




 89%|████████▉ | 44710/50000 [12:13<01:53, 46.52it/s]




 89%|████████▉ | 44715/50000 [12:14<02:07, 41.47it/s]




 89%|████████▉ | 44720/50000 [12:14<02:02, 43.17it/s]




 89%|████████▉ | 44725/50000 [12:14<01:58, 44.69it/s]




 89%|████████▉ | 44730/50000 [12:14<01:56, 45.07it/s]




 89%|████████▉ | 44735/50000 [12:14<01:55, 45.78it/s]




 89%|████████▉ | 44740/50000 [12:14<01:54, 46.03it/s]




 89%|████████▉ | 44746/50000 [12:14<01:49, 48.15it/s]




 90%|████████▉ | 44751/50000 [12:14<01:48, 48.31it/s]




 90%|████████▉ | 44756/50000 [12:14<01:51, 47.19it/s]




 90%|████████▉ | 44761/50000 [12:15<01:49, 47.90it/s]




 90%|████████▉ | 44769/50000 [12:15<01:38, 52.86it/s]




 90%|████████▉ | 44775/50000 [12:15<01:37, 53.69it/s]




 90%|████████▉ | 44782/50000 [12:15<01:31, 57.19it/s]




 90%|████████▉ | 44788/50000 [12:15<01:32, 56.26it/s]




 90%|████████▉ | 44795/50000 [12:15<01:27, 59.44it/s]




 90%|████████▉ | 44803/50000 [12:15<01:21, 63.53it/s]




 90%|████████▉ | 44810/50000 [12:15<01:21, 63.61it/s]




 90%|████████▉ | 44817/50000 [12:15<01:19, 64.91it/s]




 90%|████████▉ | 44824/50000 [12:15<01:21, 63.50it/s]




 90%|████████▉ | 44831/50000 [12:16<01:29, 57.99it/s]




 90%|████████▉ | 44838/50000 [12:16<01:30, 57.26it/s]




 90%|████████▉ | 44845/50000 [12:16<01:30, 57.27it/s]




 90%|████████▉ | 44851/50000 [12:16<01:37, 53.04it/s]




 90%|████████▉ | 44857/50000 [12:16<01:39, 51.59it/s]




 90%|████████▉ | 44863/50000 [12:16<01:39, 51.54it/s]




 90%|████████▉ | 44869/50000 [12:16<01:54, 44.93it/s]




 90%|████████▉ | 44875/50000 [12:17<01:49, 46.91it/s]




 90%|████████▉ | 44883/50000 [12:17<01:37, 52.23it/s]




 90%|████████▉ | 44891/50000 [12:17<01:28, 57.95it/s]




 90%|████████▉ | 44898/50000 [12:17<01:24, 60.05it/s]




 90%|████████▉ | 44905/50000 [12:17<01:25, 59.73it/s]




 90%|████████▉ | 44912/50000 [12:17<01:27, 58.17it/s]




 90%|████████▉ | 44918/50000 [12:17<01:26, 58.66it/s]




 90%|████████▉ | 44924/50000 [12:17<01:26, 58.68it/s]




 90%|████████▉ | 44931/50000 [12:17<01:22, 61.39it/s]




 90%|████████▉ | 44938/50000 [12:18<01:20, 62.93it/s]




 90%|████████▉ | 44945/50000 [12:18<01:18, 64.23it/s]




 90%|████████▉ | 44952/50000 [12:18<01:17, 64.81it/s]




 90%|████████▉ | 44959/50000 [12:18<01:20, 62.78it/s]




 90%|████████▉ | 44967/50000 [12:18<01:16, 65.51it/s]




 90%|████████▉ | 44975/50000 [12:18<01:14, 67.74it/s]




 90%|████████▉ | 44982/50000 [12:18<01:17, 64.50it/s]




 90%|████████▉ | 44989/50000 [12:18<01:18, 63.76it/s]




 90%|████████▉ | 44996/50000 [12:18<01:17, 64.48it/s]




 90%|█████████ | 45004/50000 [12:19<01:15, 66.47it/s]




 90%|█████████ | 45011/50000 [12:19<01:20, 62.16it/s]




 90%|█████████ | 45018/50000 [12:19<01:21, 61.10it/s]




 90%|█████████ | 45025/50000 [12:19<01:23, 59.45it/s]




 90%|█████████ | 45032/50000 [12:19<01:24, 58.86it/s]




 90%|█████████ | 45039/50000 [12:19<01:20, 61.53it/s]




 90%|█████████ | 45046/50000 [12:19<01:19, 62.69it/s]




 90%|█████████ | 45053/50000 [12:19<01:22, 60.26it/s]




 90%|█████████ | 45060/50000 [12:19<01:19, 62.10it/s]




 90%|█████████ | 45067/50000 [12:20<01:17, 63.80it/s]




 90%|█████████ | 45074/50000 [12:20<01:17, 63.63it/s]




 90%|█████████ | 45082/50000 [12:20<01:12, 67.49it/s]




 90%|█████████ | 45089/50000 [12:20<01:14, 65.78it/s]




 90%|█████████ | 45096/50000 [12:20<01:16, 63.93it/s]




 90%|█████████ | 45103/50000 [12:20<01:22, 59.50it/s]




 90%|█████████ | 45110/50000 [12:20<01:21, 59.72it/s]




 90%|█████████ | 45117/50000 [12:20<01:20, 60.97it/s]




 90%|█████████ | 45124/50000 [12:20<01:19, 61.55it/s]




 90%|█████████ | 45131/50000 [12:21<01:21, 59.59it/s]




 90%|█████████ | 45138/50000 [12:21<01:24, 57.51it/s]




 90%|█████████ | 45146/50000 [12:21<01:18, 62.12it/s]




 90%|█████████ | 45153/50000 [12:21<01:20, 60.35it/s]




 90%|█████████ | 45161/50000 [12:21<01:15, 63.95it/s]




 90%|█████████ | 45169/50000 [12:21<01:12, 66.56it/s]




 90%|█████████ | 45176/50000 [12:21<01:13, 65.53it/s]




 90%|█████████ | 45183/50000 [12:21<01:14, 64.93it/s]




 90%|█████████ | 45190/50000 [12:22<01:16, 63.27it/s]




 90%|█████████ | 45197/50000 [12:22<01:21, 58.66it/s]




 90%|█████████ | 45204/50000 [12:22<01:18, 60.71it/s]




 90%|█████████ | 45211/50000 [12:22<01:18, 61.23it/s]




 90%|█████████ | 45218/50000 [12:22<01:20, 59.36it/s]




 90%|█████████ | 45225/50000 [12:22<01:17, 61.43it/s]




 90%|█████████ | 45232/50000 [12:22<01:15, 62.96it/s]




 90%|█████████ | 45239/50000 [12:22<01:14, 63.74it/s]




 90%|█████████ | 45247/50000 [12:22<01:11, 66.74it/s]




 91%|█████████ | 45254/50000 [12:23<01:14, 63.59it/s]




 91%|█████████ | 45261/50000 [12:23<01:13, 64.53it/s]




 91%|█████████ | 45268/50000 [12:23<01:16, 62.09it/s]




 91%|█████████ | 45275/50000 [12:23<01:19, 59.72it/s]




 91%|█████████ | 45282/50000 [12:23<01:22, 57.45it/s]




 91%|█████████ | 45288/50000 [12:23<01:24, 55.65it/s]




 91%|█████████ | 45294/50000 [12:23<01:25, 54.91it/s]




 91%|█████████ | 45300/50000 [12:23<01:26, 54.10it/s]




 91%|█████████ | 45306/50000 [12:23<01:29, 52.64it/s]




 91%|█████████ | 45312/50000 [12:24<01:34, 49.43it/s]




 91%|█████████ | 45319/50000 [12:24<01:28, 52.89it/s]




 91%|█████████ | 45327/50000 [12:24<01:21, 57.58it/s]




 91%|█████████ | 45334/50000 [12:24<01:18, 59.77it/s]




 91%|█████████ | 45341/50000 [12:24<01:16, 60.68it/s]




 91%|█████████ | 45348/50000 [12:24<01:13, 63.17it/s]




 91%|█████████ | 45355/50000 [12:24<01:14, 62.34it/s]




 91%|█████████ | 45363/50000 [12:24<01:10, 65.66it/s]




 91%|█████████ | 45370/50000 [12:25<01:11, 64.91it/s]




 91%|█████████ | 45377/50000 [12:25<01:12, 63.87it/s]




 91%|█████████ | 45384/50000 [12:25<01:13, 62.49it/s]




 91%|█████████ | 45391/50000 [12:25<01:11, 64.44it/s]




 91%|█████████ | 45398/50000 [12:25<01:12, 63.72it/s]




 91%|█████████ | 45405/50000 [12:25<01:14, 61.48it/s]




 91%|█████████ | 45412/50000 [12:25<01:13, 62.74it/s]




 91%|█████████ | 45419/50000 [12:25<01:11, 63.92it/s]




 91%|█████████ | 45426/50000 [12:25<01:18, 58.26it/s]




 91%|█████████ | 45432/50000 [12:26<01:17, 58.64it/s]




 91%|█████████ | 45439/50000 [12:26<01:16, 59.81it/s]




 91%|█████████ | 45446/50000 [12:26<01:18, 58.22it/s]




 91%|█████████ | 45454/50000 [12:26<01:14, 60.63it/s]




 91%|█████████ | 45461/50000 [12:26<01:17, 58.92it/s]




 91%|█████████ | 45468/50000 [12:26<01:13, 61.57it/s]




 91%|█████████ | 45475/50000 [12:26<01:18, 57.77it/s]




 91%|█████████ | 45482/50000 [12:26<01:14, 60.85it/s]




 91%|█████████ | 45489/50000 [12:26<01:12, 62.36it/s]




 91%|█████████ | 45496/50000 [12:27<01:17, 58.40it/s]




 91%|█████████ | 45502/50000 [12:27<01:20, 55.80it/s]




 91%|█████████ | 45508/50000 [12:27<01:22, 54.71it/s]




 91%|█████████ | 45514/50000 [12:27<01:20, 55.46it/s]




 91%|█████████ | 45520/50000 [12:27<01:22, 54.62it/s]




 91%|█████████ | 45526/50000 [12:27<01:20, 55.78it/s]




 91%|█████████ | 45532/50000 [12:27<01:23, 53.31it/s]




 91%|█████████ | 45539/50000 [12:27<01:19, 55.87it/s]




 91%|█████████ | 45546/50000 [12:27<01:15, 58.99it/s]




 91%|█████████ | 45552/50000 [12:28<01:17, 57.30it/s]




 91%|█████████ | 45558/50000 [12:28<01:19, 55.55it/s]




 91%|█████████ | 45564/50000 [12:28<01:22, 53.95it/s]




 91%|█████████ | 45571/50000 [12:28<01:17, 57.13it/s]




 91%|█████████ | 45579/50000 [12:28<01:11, 61.95it/s]




 91%|█████████ | 45586/50000 [12:28<01:10, 62.49it/s]




 91%|█████████ | 45593/50000 [12:28<01:09, 63.22it/s]




 91%|█████████ | 45600/50000 [12:28<01:08, 64.35it/s]




 91%|█████████ | 45607/50000 [12:28<01:07, 64.72it/s]




 91%|█████████ | 45614/50000 [12:29<01:09, 62.88it/s]




 91%|█████████ | 45621/50000 [12:29<01:12, 60.23it/s]




 91%|█████████▏| 45629/50000 [12:29<01:07, 64.62it/s]




 91%|█████████▏| 45636/50000 [12:29<01:06, 66.01it/s]




 91%|█████████▏| 45643/50000 [12:29<01:08, 63.21it/s]




 91%|█████████▏| 45650/50000 [12:29<01:14, 58.48it/s]




 91%|█████████▏| 45656/50000 [12:29<01:15, 57.78it/s]




 91%|█████████▏| 45662/50000 [12:29<01:23, 51.83it/s]




 91%|█████████▏| 45668/50000 [12:30<01:26, 49.90it/s]




 91%|█████████▏| 45674/50000 [12:30<01:29, 48.40it/s]




 91%|█████████▏| 45679/50000 [12:30<01:32, 46.59it/s]




 91%|█████████▏| 45685/50000 [12:30<01:27, 49.12it/s]




 91%|█████████▏| 45691/50000 [12:30<01:27, 49.35it/s]




 91%|█████████▏| 45697/50000 [12:30<01:30, 47.30it/s]




 91%|█████████▏| 45703/50000 [12:30<01:25, 50.04it/s]




 91%|█████████▏| 45709/50000 [12:30<01:24, 50.70it/s]




 91%|█████████▏| 45715/50000 [12:31<01:27, 49.21it/s]




 91%|█████████▏| 45722/50000 [12:31<01:19, 53.61it/s]




 91%|█████████▏| 45728/50000 [12:31<01:18, 54.48it/s]




 91%|█████████▏| 45736/50000 [12:31<01:11, 59.68it/s]




 91%|█████████▏| 45743/50000 [12:31<01:12, 58.87it/s]




 92%|█████████▏| 45750/50000 [12:31<01:09, 61.21it/s]




 92%|█████████▏| 45758/50000 [12:31<01:05, 64.62it/s]




 92%|█████████▏| 45765/50000 [12:31<01:05, 65.09it/s]




 92%|█████████▏| 45772/50000 [12:31<01:06, 63.47it/s]




 92%|█████████▏| 45779/50000 [12:32<01:05, 64.27it/s]




 92%|█████████▏| 45786/50000 [12:32<01:09, 60.96it/s]




 92%|█████████▏| 45793/50000 [12:32<01:09, 60.69it/s]




 92%|█████████▏| 45800/50000 [12:32<01:08, 61.11it/s]




 92%|█████████▏| 45807/50000 [12:32<01:09, 59.91it/s]




 92%|█████████▏| 45814/50000 [12:32<01:12, 57.63it/s]




 92%|█████████▏| 45820/50000 [12:32<01:15, 55.01it/s]




 92%|█████████▏| 45826/50000 [12:32<01:15, 55.06it/s]




 92%|█████████▏| 45832/50000 [12:32<01:14, 55.71it/s]




 92%|█████████▏| 45838/50000 [12:33<01:16, 54.64it/s]




 92%|█████████▏| 45845/50000 [12:33<01:13, 56.55it/s]




 92%|█████████▏| 45852/50000 [12:33<01:10, 58.76it/s]




 92%|█████████▏| 45858/50000 [12:33<01:10, 58.66it/s]




 92%|█████████▏| 45865/50000 [12:33<01:08, 60.42it/s]




 92%|█████████▏| 45872/50000 [12:33<01:07, 61.08it/s]




 92%|█████████▏| 45879/50000 [12:33<01:08, 60.12it/s]




 92%|█████████▏| 45886/50000 [12:33<01:05, 62.66it/s]




 92%|█████████▏| 45893/50000 [12:33<01:03, 64.21it/s]




 92%|█████████▏| 45901/50000 [12:34<01:01, 66.76it/s]




 92%|█████████▏| 45908/50000 [12:34<01:01, 66.31it/s]




 92%|█████████▏| 45915/50000 [12:34<01:02, 65.63it/s]




 92%|█████████▏| 45922/50000 [12:34<01:05, 61.81it/s]




 92%|█████████▏| 45930/50000 [12:34<01:03, 64.15it/s]




 92%|█████████▏| 45938/50000 [12:34<01:00, 66.88it/s]




 92%|█████████▏| 45945/50000 [12:34<01:00, 66.58it/s]




 92%|█████████▏| 45952/50000 [12:34<01:02, 65.08it/s]




 92%|█████████▏| 45960/50000 [12:34<00:59, 68.02it/s]




 92%|█████████▏| 45967/50000 [12:35<01:04, 62.26it/s]




 92%|█████████▏| 45974/50000 [12:35<01:02, 64.09it/s]




 92%|█████████▏| 45981/50000 [12:35<01:06, 60.84it/s]




 92%|█████████▏| 45988/50000 [12:35<01:11, 55.81it/s]




 92%|█████████▏| 45995/50000 [12:35<01:08, 58.87it/s]




 92%|█████████▏| 46002/50000 [12:35<01:05, 60.58it/s]




 92%|█████████▏| 46009/50000 [12:35<01:05, 61.19it/s]




 92%|█████████▏| 46017/50000 [12:35<01:01, 64.68it/s]




 92%|█████████▏| 46024/50000 [12:35<01:03, 63.03it/s]




 92%|█████████▏| 46031/50000 [12:36<01:04, 61.43it/s]




 92%|█████████▏| 46038/50000 [12:36<01:06, 59.35it/s]




 92%|█████████▏| 46045/50000 [12:36<01:05, 60.77it/s]




 92%|█████████▏| 46052/50000 [12:36<01:03, 61.73it/s]




 92%|█████████▏| 46061/50000 [12:36<00:59, 65.66it/s]




 92%|█████████▏| 46068/50000 [12:36<01:00, 65.10it/s]




 92%|█████████▏| 46075/50000 [12:36<01:07, 58.40it/s]




 92%|█████████▏| 46082/50000 [12:36<01:09, 56.59it/s]




 92%|█████████▏| 46088/50000 [12:37<01:11, 54.64it/s]




 92%|█████████▏| 46095/50000 [12:37<01:09, 56.55it/s]




 92%|█████████▏| 46101/50000 [12:37<01:10, 55.67it/s]




 92%|█████████▏| 46108/50000 [12:37<01:06, 58.39it/s]




 92%|█████████▏| 46115/50000 [12:37<01:03, 60.85it/s]




 92%|█████████▏| 46122/50000 [12:37<01:02, 62.03it/s]




 92%|█████████▏| 46129/50000 [12:37<01:02, 61.57it/s]




 92%|█████████▏| 46137/50000 [12:37<00:59, 65.21it/s]




 92%|█████████▏| 46144/50000 [12:37<01:04, 59.59it/s]




 92%|█████████▏| 46151/50000 [12:38<01:09, 55.07it/s]




 92%|█████████▏| 46157/50000 [12:38<01:09, 54.95it/s]




 92%|█████████▏| 46164/50000 [12:38<01:05, 58.49it/s]




 92%|█████████▏| 46172/50000 [12:38<01:01, 62.33it/s]




 92%|█████████▏| 46179/50000 [12:38<01:01, 61.77it/s]




 92%|█████████▏| 46186/50000 [12:38<00:59, 63.90it/s]




 92%|█████████▏| 46193/50000 [12:38<01:01, 61.85it/s]




 92%|█████████▏| 46200/50000 [12:38<01:03, 59.67it/s]




 92%|█████████▏| 46207/50000 [12:39<01:05, 58.27it/s]




 92%|█████████▏| 46215/50000 [12:39<01:01, 61.72it/s]




 92%|█████████▏| 46222/50000 [12:39<01:01, 61.84it/s]




 92%|█████████▏| 46229/50000 [12:39<01:02, 60.02it/s]




 92%|█████████▏| 46236/50000 [12:39<01:04, 58.08it/s]




 92%|█████████▏| 46242/50000 [12:39<01:11, 52.74it/s]




 92%|█████████▏| 46248/50000 [12:39<01:09, 53.81it/s]




 93%|█████████▎| 46255/50000 [12:39<01:05, 57.58it/s]




 93%|█████████▎| 46261/50000 [12:39<01:05, 57.49it/s]




 93%|█████████▎| 46267/50000 [12:40<01:05, 56.78it/s]




 93%|█████████▎| 46274/50000 [12:40<01:02, 59.77it/s]




 93%|█████████▎| 46281/50000 [12:40<01:03, 58.20it/s]




 93%|█████████▎| 46288/50000 [12:40<01:01, 60.08it/s]




 93%|█████████▎| 46296/50000 [12:40<00:57, 64.66it/s]




 93%|█████████▎| 46303/50000 [12:40<01:00, 60.64it/s]




 93%|█████████▎| 46310/50000 [12:40<01:01, 60.14it/s]




 93%|█████████▎| 46317/50000 [12:40<00:59, 62.17it/s]




 93%|█████████▎| 46324/50000 [12:41<01:00, 61.02it/s]




 93%|█████████▎| 46331/50000 [12:41<00:58, 62.82it/s]




 93%|█████████▎| 46339/50000 [12:41<00:55, 65.71it/s]




 93%|█████████▎| 46346/50000 [12:41<00:59, 61.85it/s]




 93%|█████████▎| 46353/50000 [12:41<00:58, 62.10it/s]




 93%|█████████▎| 46361/50000 [12:41<00:55, 65.15it/s]




 93%|█████████▎| 46368/50000 [12:41<00:57, 63.68it/s]




 93%|█████████▎| 46376/50000 [12:41<00:54, 66.04it/s]




 93%|█████████▎| 46383/50000 [12:41<00:57, 62.72it/s]




 93%|█████████▎| 46390/50000 [12:42<01:02, 57.47it/s]




 93%|█████████▎| 46396/50000 [12:42<01:02, 57.77it/s]




 93%|█████████▎| 46402/50000 [12:42<01:02, 57.80it/s]




 93%|█████████▎| 46410/50000 [12:42<00:58, 61.21it/s]




 93%|█████████▎| 46417/50000 [12:42<00:58, 60.84it/s]




 93%|█████████▎| 46424/50000 [12:42<00:59, 60.50it/s]




 93%|█████████▎| 46432/50000 [12:42<00:55, 64.78it/s]




 93%|█████████▎| 46439/50000 [12:42<00:56, 63.01it/s]




 93%|█████████▎| 46447/50000 [12:42<00:54, 65.57it/s]




 93%|█████████▎| 46455/50000 [12:43<00:51, 69.19it/s]




 93%|█████████▎| 46463/50000 [12:43<00:51, 68.30it/s]




 93%|█████████▎| 46470/50000 [12:43<00:51, 68.65it/s]




 93%|█████████▎| 46477/50000 [12:43<00:54, 64.53it/s]




 93%|█████████▎| 46484/50000 [12:43<00:53, 65.21it/s]




 93%|█████████▎| 46491/50000 [12:43<00:53, 65.51it/s]




 93%|█████████▎| 46498/50000 [12:43<00:52, 66.09it/s]




 93%|█████████▎| 46506/50000 [12:43<00:52, 67.15it/s]




 93%|█████████▎| 46513/50000 [12:43<00:52, 65.92it/s]




 93%|█████████▎| 46521/50000 [12:44<00:50, 68.22it/s]




 93%|█████████▎| 46528/50000 [12:44<00:52, 66.64it/s]




 93%|█████████▎| 46535/50000 [12:44<00:52, 66.60it/s]




 93%|█████████▎| 46542/50000 [12:44<00:52, 66.38it/s]




 93%|█████████▎| 46549/50000 [12:44<00:54, 62.77it/s]




 93%|█████████▎| 46556/50000 [12:44<00:58, 58.72it/s]




 93%|█████████▎| 46562/50000 [12:44<00:58, 58.80it/s]




 93%|█████████▎| 46569/50000 [12:44<00:56, 60.37it/s]




 93%|█████████▎| 46577/50000 [12:44<00:53, 63.51it/s]




 93%|█████████▎| 46584/50000 [12:45<00:57, 59.86it/s]




 93%|█████████▎| 46591/50000 [12:45<00:54, 62.12it/s]




 93%|█████████▎| 46598/50000 [12:45<00:59, 57.54it/s]




 93%|█████████▎| 46605/50000 [12:45<00:56, 60.20it/s]




 93%|█████████▎| 46612/50000 [12:45<01:00, 55.96it/s]




 93%|█████████▎| 46618/50000 [12:45<01:05, 51.71it/s]




 93%|█████████▎| 46624/50000 [12:45<01:08, 49.09it/s]




 93%|█████████▎| 46630/50000 [12:45<01:10, 47.75it/s]




 93%|█████████▎| 46635/50000 [12:46<01:13, 45.48it/s]




 93%|█████████▎| 46640/50000 [12:46<01:19, 42.33it/s]




 93%|█████████▎| 46645/50000 [12:46<01:16, 44.00it/s]




 93%|█████████▎| 46650/50000 [12:46<01:13, 45.30it/s]




 93%|█████████▎| 46655/50000 [12:46<01:12, 46.14it/s]




 93%|█████████▎| 46660/50000 [12:46<01:11, 46.61it/s]




 93%|█████████▎| 46665/50000 [12:46<01:14, 44.68it/s]




 93%|█████████▎| 46670/50000 [12:46<01:22, 40.39it/s]




 93%|█████████▎| 46675/50000 [12:47<01:22, 40.20it/s]




 93%|█████████▎| 46680/50000 [12:47<01:19, 41.78it/s]




 93%|█████████▎| 46686/50000 [12:47<01:12, 45.42it/s]




 93%|█████████▎| 46691/50000 [12:47<01:13, 45.07it/s]




 93%|█████████▎| 46696/50000 [12:47<01:15, 43.85it/s]




 93%|█████████▎| 46701/50000 [12:47<01:14, 44.47it/s]




 93%|█████████▎| 46709/50000 [12:47<01:04, 50.65it/s]




 93%|█████████▎| 46715/50000 [12:47<01:03, 51.52it/s]




 93%|█████████▎| 46722/50000 [12:47<00:59, 55.00it/s]




 93%|█████████▎| 46728/50000 [12:48<00:59, 54.83it/s]




 93%|█████████▎| 46735/50000 [12:48<00:57, 57.17it/s]




 93%|█████████▎| 46741/50000 [12:48<01:03, 51.48it/s]




 93%|█████████▎| 46747/50000 [12:48<01:03, 51.48it/s]




 94%|█████████▎| 46753/50000 [12:48<01:01, 52.54it/s]




 94%|█████████▎| 46759/50000 [12:48<01:11, 45.56it/s]




 94%|█████████▎| 46765/50000 [12:48<01:09, 46.84it/s]




 94%|█████████▎| 46770/50000 [12:48<01:14, 43.10it/s]




 94%|█████████▎| 46775/50000 [12:49<01:14, 43.19it/s]




 94%|█████████▎| 46781/50000 [12:49<01:09, 46.63it/s]




 94%|█████████▎| 46787/50000 [12:49<01:06, 48.55it/s]




 94%|█████████▎| 46793/50000 [12:49<01:03, 50.37it/s]




 94%|█████████▎| 46800/50000 [12:49<00:59, 53.83it/s]




 94%|█████████▎| 46806/50000 [12:49<00:58, 54.16it/s]




 94%|█████████▎| 46813/50000 [12:49<00:55, 57.92it/s]




 94%|█████████▎| 46820/50000 [12:49<00:53, 59.64it/s]




 94%|█████████▎| 46828/50000 [12:49<00:49, 63.54it/s]




 94%|█████████▎| 46835/50000 [12:50<00:51, 61.29it/s]




 94%|█████████▎| 46842/50000 [12:50<00:52, 60.42it/s]




 94%|█████████▎| 46849/50000 [12:50<00:51, 60.76it/s]




 94%|█████████▎| 46856/50000 [12:50<00:50, 61.81it/s]




 94%|█████████▎| 46863/50000 [12:50<00:50, 61.90it/s]




 94%|█████████▎| 46871/50000 [12:50<00:49, 63.75it/s]




 94%|█████████▍| 46878/50000 [12:50<00:50, 61.75it/s]




 94%|█████████▍| 46885/50000 [12:50<00:49, 63.03it/s]




 94%|█████████▍| 46892/50000 [12:50<00:51, 60.79it/s]




 94%|█████████▍| 46899/50000 [12:51<00:50, 61.86it/s]




 94%|█████████▍| 46906/50000 [12:51<00:49, 61.94it/s]




 94%|█████████▍| 46913/50000 [12:51<00:53, 57.57it/s]




 94%|█████████▍| 46919/50000 [12:51<00:53, 57.98it/s]




 94%|█████████▍| 46925/50000 [12:51<00:52, 58.11it/s]




 94%|█████████▍| 46932/50000 [12:51<00:50, 60.33it/s]




 94%|█████████▍| 46939/50000 [12:51<00:53, 57.42it/s]




 94%|█████████▍| 46947/50000 [12:51<00:49, 61.61it/s]




 94%|█████████▍| 46954/50000 [12:52<00:50, 60.80it/s]




 94%|█████████▍| 46961/50000 [12:52<00:50, 60.40it/s]




 94%|█████████▍| 46969/50000 [12:52<00:46, 64.96it/s]




 94%|█████████▍| 46977/50000 [12:52<00:44, 68.18it/s]




 94%|█████████▍| 46984/50000 [12:52<00:43, 68.57it/s]




 94%|█████████▍| 46991/50000 [12:52<00:45, 66.69it/s]




 94%|█████████▍| 46998/50000 [12:52<00:44, 66.83it/s]




 94%|█████████▍| 47005/50000 [12:52<00:44, 66.83it/s]




 94%|█████████▍| 47013/50000 [12:52<00:43, 68.90it/s]




 94%|█████████▍| 47022/50000 [12:52<00:41, 72.21it/s]




 94%|█████████▍| 47030/50000 [12:53<00:43, 68.35it/s]




 94%|█████████▍| 47038/50000 [12:53<00:42, 69.66it/s]




 94%|█████████▍| 47046/50000 [12:53<00:44, 67.06it/s]




 94%|█████████▍| 47053/50000 [12:53<00:44, 66.62it/s]




 94%|█████████▍| 47060/50000 [12:53<00:45, 64.30it/s]




 94%|█████████▍| 47067/50000 [12:53<00:45, 64.51it/s]




 94%|█████████▍| 47074/50000 [12:53<00:45, 64.83it/s]




 94%|█████████▍| 47081/50000 [12:53<00:45, 64.16it/s]




 94%|█████████▍| 47088/50000 [12:54<00:49, 58.60it/s]




 94%|█████████▍| 47094/50000 [12:54<00:52, 55.78it/s]




 94%|█████████▍| 47101/50000 [12:54<00:49, 59.14it/s]




 94%|█████████▍| 47108/50000 [12:54<00:47, 60.94it/s]




 94%|█████████▍| 47115/50000 [12:54<00:48, 59.42it/s]




 94%|█████████▍| 47122/50000 [12:54<00:47, 60.51it/s]




 94%|█████████▍| 47130/50000 [12:54<00:45, 63.02it/s]




 94%|█████████▍| 47137/50000 [12:54<00:46, 61.91it/s]




 94%|█████████▍| 47145/50000 [12:54<00:43, 65.48it/s]




 94%|█████████▍| 47152/50000 [12:55<00:44, 64.43it/s]




 94%|█████████▍| 47159/50000 [12:55<00:43, 65.50it/s]




 94%|█████████▍| 47168/50000 [12:55<00:40, 70.21it/s]




 94%|█████████▍| 47176/50000 [12:55<00:42, 66.80it/s]




 94%|█████████▍| 47184/50000 [12:55<00:41, 68.35it/s]




 94%|█████████▍| 47191/50000 [12:55<00:43, 65.15it/s]




 94%|█████████▍| 47199/50000 [12:55<00:41, 67.13it/s]




 94%|█████████▍| 47206/50000 [12:55<00:43, 64.28it/s]




 94%|█████████▍| 47213/50000 [12:55<00:44, 62.10it/s]




 94%|█████████▍| 47220/50000 [12:56<00:45, 60.81it/s]




 94%|█████████▍| 47227/50000 [12:56<00:45, 61.52it/s]




 94%|█████████▍| 47235/50000 [12:56<00:42, 65.17it/s]




 94%|█████████▍| 47242/50000 [12:56<00:42, 64.76it/s]




 94%|█████████▍| 47249/50000 [12:56<00:42, 64.83it/s]




 95%|█████████▍| 47256/50000 [12:56<00:44, 61.14it/s]




 95%|█████████▍| 47263/50000 [12:56<00:46, 58.43it/s]




 95%|█████████▍| 47270/50000 [12:56<00:45, 60.10it/s]




 95%|█████████▍| 47277/50000 [12:57<00:47, 57.77it/s]




 95%|█████████▍| 47284/50000 [12:57<00:45, 59.23it/s]




 95%|█████████▍| 47290/50000 [12:57<00:48, 55.86it/s]




 95%|█████████▍| 47296/50000 [12:57<00:51, 52.73it/s]




 95%|█████████▍| 47303/50000 [12:57<00:47, 56.30it/s]




 95%|█████████▍| 47310/50000 [12:57<00:46, 57.92it/s]




 95%|█████████▍| 47316/50000 [12:57<00:47, 56.43it/s]




 95%|█████████▍| 47322/50000 [12:57<00:48, 54.83it/s]




 95%|█████████▍| 47329/50000 [12:57<00:45, 58.24it/s]




 95%|█████████▍| 47336/50000 [12:58<00:45, 58.76it/s]




 95%|█████████▍| 47342/50000 [12:58<00:47, 56.26it/s]




 95%|█████████▍| 47348/50000 [12:58<00:46, 57.04it/s]




 95%|█████████▍| 47355/50000 [12:58<00:44, 59.06it/s]




 95%|█████████▍| 47361/50000 [12:58<00:45, 57.68it/s]




 95%|█████████▍| 47367/50000 [12:58<00:45, 57.40it/s]




 95%|█████████▍| 47376/50000 [12:58<00:41, 62.82it/s]




 95%|█████████▍| 47383/50000 [12:58<00:41, 62.77it/s]




 95%|█████████▍| 47390/50000 [12:58<00:40, 63.88it/s]




 95%|█████████▍| 47397/50000 [12:59<00:40, 64.30it/s]




 95%|█████████▍| 47404/50000 [12:59<00:41, 62.77it/s]




 95%|█████████▍| 47412/50000 [12:59<00:39, 65.03it/s]




 95%|█████████▍| 47420/50000 [12:59<00:38, 67.21it/s]




 95%|█████████▍| 47428/50000 [12:59<00:37, 68.48it/s]




 95%|█████████▍| 47435/50000 [12:59<00:39, 64.24it/s]




 95%|█████████▍| 47442/50000 [12:59<00:40, 62.57it/s]




 95%|█████████▍| 47449/50000 [12:59<00:40, 62.27it/s]




 95%|█████████▍| 47457/50000 [12:59<00:39, 65.12it/s]




 95%|█████████▍| 47464/50000 [13:00<00:41, 60.53it/s]




 95%|█████████▍| 47473/50000 [13:00<00:38, 66.13it/s]




 95%|█████████▍| 47480/50000 [13:00<00:39, 64.10it/s]




 95%|█████████▍| 47487/50000 [13:00<00:40, 61.82it/s]




 95%|█████████▍| 47494/50000 [13:00<00:39, 63.93it/s]




 95%|█████████▌| 47501/50000 [13:00<00:38, 64.42it/s]




 95%|█████████▌| 47508/50000 [13:00<00:38, 65.31it/s]




 95%|█████████▌| 47515/50000 [13:00<00:37, 66.51it/s]




 95%|█████████▌| 47523/50000 [13:00<00:36, 67.97it/s]




 95%|█████████▌| 47530/50000 [13:01<00:36, 68.02it/s]




 95%|█████████▌| 47537/50000 [13:01<00:37, 65.58it/s]




 95%|█████████▌| 47544/50000 [13:01<00:38, 63.11it/s]




 95%|█████████▌| 47551/50000 [13:01<00:40, 59.76it/s]




 95%|█████████▌| 47558/50000 [13:01<00:45, 53.65it/s]




 95%|█████████▌| 47564/50000 [13:01<00:46, 52.75it/s]




 95%|█████████▌| 47570/50000 [13:01<00:45, 53.39it/s]




 95%|█████████▌| 47576/50000 [13:01<00:47, 51.10it/s]




 95%|█████████▌| 47582/50000 [13:02<00:50, 48.18it/s]




 95%|█████████▌| 47587/50000 [13:02<00:50, 48.19it/s]




 95%|█████████▌| 47593/50000 [13:02<00:49, 49.11it/s]




 95%|█████████▌| 47598/50000 [13:02<00:51, 46.66it/s]




 95%|█████████▌| 47603/50000 [13:02<00:54, 43.78it/s]




 95%|█████████▌| 47608/50000 [13:02<00:55, 43.05it/s]




 95%|█████████▌| 47613/50000 [13:02<00:54, 43.55it/s]




 95%|█████████▌| 47618/50000 [13:02<00:53, 44.79it/s]




 95%|█████████▌| 47623/50000 [13:03<00:52, 45.02it/s]




 95%|█████████▌| 47628/50000 [13:03<00:54, 43.54it/s]




 95%|█████████▌| 47634/50000 [13:03<00:50, 46.59it/s]




 95%|█████████▌| 47641/50000 [13:03<00:46, 51.14it/s]




 95%|█████████▌| 47648/50000 [13:03<00:43, 53.72it/s]




 95%|█████████▌| 47654/50000 [13:03<00:42, 54.89it/s]




 95%|█████████▌| 47661/50000 [13:03<00:39, 58.59it/s]




 95%|█████████▌| 47668/50000 [13:03<00:38, 61.32it/s]




 95%|█████████▌| 47675/50000 [13:03<00:40, 57.47it/s]




 95%|█████████▌| 47682/50000 [13:04<00:38, 60.46it/s]




 95%|█████████▌| 47689/50000 [13:04<00:38, 60.79it/s]




 95%|█████████▌| 47696/50000 [13:04<00:40, 56.46it/s]




 95%|█████████▌| 47702/50000 [13:04<00:41, 54.98it/s]




 95%|█████████▌| 47708/50000 [13:04<00:46, 48.81it/s]




 95%|█████████▌| 47714/50000 [13:04<00:47, 48.43it/s]




 95%|█████████▌| 47719/50000 [13:04<00:47, 48.08it/s]




 95%|█████████▌| 47725/50000 [13:04<00:46, 48.68it/s]




 95%|█████████▌| 47730/50000 [13:05<00:49, 45.87it/s]




 95%|█████████▌| 47735/50000 [13:05<00:50, 44.56it/s]




 95%|█████████▌| 47740/50000 [13:05<00:52, 43.07it/s]




 95%|█████████▌| 47745/50000 [13:05<00:51, 43.91it/s]




 96%|█████████▌| 47751/50000 [13:05<00:48, 46.67it/s]




 96%|█████████▌| 47758/50000 [13:05<00:43, 51.31it/s]




 96%|█████████▌| 47764/50000 [13:05<00:45, 49.45it/s]




 96%|█████████▌| 47771/50000 [13:05<00:41, 53.15it/s]




 96%|█████████▌| 47779/50000 [13:05<00:38, 57.48it/s]




 96%|█████████▌| 47785/50000 [13:06<00:39, 55.52it/s]




 96%|█████████▌| 47791/50000 [13:06<00:41, 53.57it/s]




 96%|█████████▌| 47797/50000 [13:06<00:41, 53.39it/s]




 96%|█████████▌| 47803/50000 [13:06<00:42, 51.83it/s]




 96%|█████████▌| 47809/50000 [13:06<00:41, 52.52it/s]




 96%|█████████▌| 47815/50000 [13:06<00:42, 50.86it/s]




 96%|█████████▌| 47822/50000 [13:06<00:39, 55.31it/s]




 96%|█████████▌| 47830/50000 [13:06<00:36, 60.03it/s]




 96%|█████████▌| 47837/50000 [13:06<00:37, 58.08it/s]




 96%|█████████▌| 47843/50000 [13:07<00:37, 57.84it/s]




 96%|█████████▌| 47849/50000 [13:07<00:37, 57.68it/s]




 96%|█████████▌| 47855/50000 [13:07<00:37, 56.59it/s]




 96%|█████████▌| 47862/50000 [13:07<00:36, 58.87it/s]




 96%|█████████▌| 47869/50000 [13:07<00:34, 61.70it/s]




 96%|█████████▌| 47877/50000 [13:07<00:32, 64.53it/s]




 96%|█████████▌| 47885/50000 [13:07<00:31, 67.04it/s]




 96%|█████████▌| 47893/50000 [13:07<00:30, 68.53it/s]




 96%|█████████▌| 47900/50000 [13:07<00:32, 64.63it/s]




 96%|█████████▌| 47907/50000 [13:08<00:32, 63.74it/s]




 96%|█████████▌| 47914/50000 [13:08<00:33, 62.15it/s]




 96%|█████████▌| 47921/50000 [13:08<00:34, 60.45it/s]




 96%|█████████▌| 47928/50000 [13:08<00:33, 62.57it/s]




 96%|█████████▌| 47935/50000 [13:08<00:33, 62.10it/s]




 96%|█████████▌| 47942/50000 [13:08<00:34, 59.57it/s]




 96%|█████████▌| 47949/50000 [13:08<00:33, 60.47it/s]




 96%|█████████▌| 47956/50000 [13:08<00:32, 62.09it/s]




 96%|█████████▌| 47964/50000 [13:08<00:31, 65.46it/s]




 96%|█████████▌| 47971/50000 [13:09<00:30, 66.43it/s]




 96%|█████████▌| 47978/50000 [13:09<00:32, 62.80it/s]




 96%|█████████▌| 47985/50000 [13:09<00:32, 61.28it/s]




 96%|█████████▌| 47992/50000 [13:09<00:32, 61.69it/s]




 96%|█████████▌| 47999/50000 [13:09<00:32, 61.33it/s]




 96%|█████████▌| 48006/50000 [13:09<00:31, 62.48it/s]




 96%|█████████▌| 48013/50000 [13:09<00:31, 62.95it/s]




 96%|█████████▌| 48020/50000 [13:09<00:31, 63.55it/s]




 96%|█████████▌| 48027/50000 [13:10<00:31, 63.45it/s]




 96%|█████████▌| 48034/50000 [13:10<00:31, 62.71it/s]




 96%|█████████▌| 48041/50000 [13:10<00:31, 62.10it/s]




 96%|█████████▌| 48048/50000 [13:10<00:36, 52.97it/s]




 96%|█████████▌| 48056/50000 [13:10<00:33, 57.46it/s]




 96%|█████████▌| 48064/50000 [13:10<00:32, 60.39it/s]




 96%|█████████▌| 48071/50000 [13:10<00:32, 58.53it/s]




 96%|█████████▌| 48079/50000 [13:10<00:30, 62.58it/s]




 96%|█████████▌| 48086/50000 [13:11<00:31, 61.05it/s]




 96%|█████████▌| 48093/50000 [13:11<00:33, 57.78it/s]




 96%|█████████▌| 48101/50000 [13:11<00:30, 62.05it/s]




 96%|█████████▌| 48108/50000 [13:11<00:29, 64.19it/s]




 96%|█████████▌| 48115/50000 [13:11<00:28, 65.27it/s]




 96%|█████████▌| 48122/50000 [13:11<00:28, 66.57it/s]




 96%|█████████▋| 48129/50000 [13:11<00:29, 63.83it/s]




 96%|█████████▋| 48138/50000 [13:11<00:27, 68.54it/s]




 96%|█████████▋| 48146/50000 [13:11<00:27, 68.02it/s]




 96%|█████████▋| 48153/50000 [13:12<00:27, 66.70it/s]




 96%|█████████▋| 48160/50000 [13:12<00:28, 64.53it/s]




 96%|█████████▋| 48168/50000 [13:12<00:27, 66.51it/s]




 96%|█████████▋| 48175/50000 [13:12<00:28, 65.13it/s]




 96%|█████████▋| 48183/50000 [13:12<00:27, 66.95it/s]




 96%|█████████▋| 48191/50000 [13:12<00:25, 69.80it/s]




 96%|█████████▋| 48199/50000 [13:12<00:27, 65.95it/s]




 96%|█████████▋| 48206/50000 [13:12<00:26, 66.59it/s]




 96%|█████████▋| 48213/50000 [13:12<00:27, 65.43it/s]




 96%|█████████▋| 48220/50000 [13:13<00:27, 65.48it/s]




 96%|█████████▋| 48227/50000 [13:13<00:27, 64.14it/s]




 96%|█████████▋| 48234/50000 [13:13<00:28, 61.68it/s]




 96%|█████████▋| 48242/50000 [13:13<00:27, 63.82it/s]




 96%|█████████▋| 48249/50000 [13:13<00:27, 63.13it/s]




 97%|█████████▋| 48256/50000 [13:13<00:31, 55.38it/s]




 97%|█████████▋| 48262/50000 [13:13<00:32, 53.13it/s]




 97%|█████████▋| 48268/50000 [13:13<00:31, 54.17it/s]




 97%|█████████▋| 48274/50000 [13:13<00:31, 55.07it/s]




 97%|█████████▋| 48281/50000 [13:14<00:30, 57.01it/s]




 97%|█████████▋| 48289/50000 [13:14<00:27, 61.99it/s]




 97%|█████████▋| 48296/50000 [13:14<00:29, 57.81it/s]




 97%|█████████▋| 48303/50000 [13:14<00:28, 59.56it/s]




 97%|█████████▋| 48311/50000 [13:14<00:26, 63.78it/s]




 97%|█████████▋| 48318/50000 [13:14<00:25, 65.21it/s]




 97%|█████████▋| 48325/50000 [13:14<00:26, 62.53it/s]




 97%|█████████▋| 48332/50000 [13:14<00:26, 64.02it/s]




 97%|█████████▋| 48340/50000 [13:14<00:24, 66.76it/s]




 97%|█████████▋| 48347/50000 [13:15<00:25, 66.03it/s]




 97%|█████████▋| 48354/50000 [13:15<00:24, 65.90it/s]




 97%|█████████▋| 48361/50000 [13:15<00:25, 65.44it/s]




 97%|█████████▋| 48368/50000 [13:15<00:25, 63.70it/s]




 97%|█████████▋| 48375/50000 [13:15<00:25, 63.22it/s]




 97%|█████████▋| 48382/50000 [13:15<00:25, 63.39it/s]




 97%|█████████▋| 48389/50000 [13:15<00:25, 62.83it/s]




 97%|█████████▋| 48396/50000 [13:15<00:25, 63.12it/s]




 97%|█████████▋| 48403/50000 [13:15<00:25, 63.28it/s]




 97%|█████████▋| 48410/50000 [13:16<00:27, 57.58it/s]




 97%|█████████▋| 48416/50000 [13:16<00:28, 54.82it/s]




 97%|█████████▋| 48423/50000 [13:16<00:27, 57.52it/s]




 97%|█████████▋| 48430/50000 [13:16<00:26, 59.27it/s]




 97%|█████████▋| 48437/50000 [13:16<00:26, 59.49it/s]




 97%|█████████▋| 48444/50000 [13:16<00:25, 60.41it/s]




 97%|█████████▋| 48453/50000 [13:16<00:23, 66.32it/s]




 97%|█████████▋| 48460/50000 [13:16<00:23, 64.46it/s]




 97%|█████████▋| 48467/50000 [13:17<00:24, 62.21it/s]




 97%|█████████▋| 48474/50000 [13:17<00:24, 61.20it/s]




 97%|█████████▋| 48481/50000 [13:17<00:24, 61.48it/s]




 97%|█████████▋| 48489/50000 [13:17<00:23, 64.98it/s]




 97%|█████████▋| 48496/50000 [13:17<00:25, 58.27it/s]




 97%|█████████▋| 48503/50000 [13:17<00:26, 56.29it/s]




 97%|█████████▋| 48509/50000 [13:17<00:28, 52.78it/s]




 97%|█████████▋| 48515/50000 [13:17<00:29, 51.03it/s]




 97%|█████████▋| 48521/50000 [13:18<00:29, 49.63it/s]




 97%|█████████▋| 48527/50000 [13:18<00:29, 49.12it/s]




 97%|█████████▋| 48532/50000 [13:18<00:30, 48.07it/s]




 97%|█████████▋| 48537/50000 [13:18<00:31, 47.09it/s]




 97%|█████████▋| 48542/50000 [13:18<00:32, 44.99it/s]




 97%|█████████▋| 48549/50000 [13:18<00:29, 49.15it/s]




 97%|█████████▋| 48555/50000 [13:18<00:31, 46.04it/s]




 97%|█████████▋| 48560/50000 [13:18<00:32, 44.67it/s]




 97%|█████████▋| 48567/50000 [13:19<00:29, 48.78it/s]




 97%|█████████▋| 48573/50000 [13:19<00:28, 50.42it/s]




 97%|█████████▋| 48581/50000 [13:19<00:25, 55.91it/s]




 97%|█████████▋| 48588/50000 [13:19<00:24, 57.28it/s]




 97%|█████████▋| 48594/50000 [13:19<00:25, 56.16it/s]




 97%|█████████▋| 48601/50000 [13:19<00:24, 58.26it/s]




 97%|█████████▋| 48608/50000 [13:19<00:23, 60.12it/s]




 97%|█████████▋| 48615/50000 [13:19<00:22, 61.74it/s]




 97%|█████████▋| 48622/50000 [13:19<00:23, 59.30it/s]




 97%|█████████▋| 48629/50000 [13:20<00:23, 57.30it/s]




 97%|█████████▋| 48635/50000 [13:20<00:24, 56.72it/s]




 97%|█████████▋| 48642/50000 [13:20<00:23, 58.38it/s]




 97%|█████████▋| 48648/50000 [13:20<00:24, 54.87it/s]




 97%|█████████▋| 48654/50000 [13:20<00:27, 49.66it/s]




 97%|█████████▋| 48660/50000 [13:20<00:27, 48.36it/s]




 97%|█████████▋| 48665/50000 [13:20<00:28, 47.22it/s]




 97%|█████████▋| 48670/50000 [13:20<00:29, 45.57it/s]




 97%|█████████▋| 48676/50000 [13:21<00:27, 48.20it/s]




 97%|█████████▋| 48683/50000 [13:21<00:25, 52.37it/s]




 97%|█████████▋| 48690/50000 [13:21<00:23, 55.61it/s]




 97%|█████████▋| 48697/50000 [13:21<00:22, 58.41it/s]




 97%|█████████▋| 48704/50000 [13:21<00:21, 60.09it/s]




 97%|█████████▋| 48711/50000 [13:21<00:20, 62.30it/s]




 97%|█████████▋| 48719/50000 [13:21<00:19, 64.36it/s]




 97%|█████████▋| 48727/50000 [13:21<00:18, 67.72it/s]




 97%|█████████▋| 48734/50000 [13:21<00:20, 62.75it/s]




 97%|█████████▋| 48742/50000 [13:21<00:19, 65.82it/s]




 97%|█████████▋| 48749/50000 [13:22<00:19, 62.58it/s]




 98%|█████████▊| 48757/50000 [13:22<00:18, 66.01it/s]




 98%|█████████▊| 48764/50000 [13:22<00:19, 64.79it/s]




 98%|█████████▊| 48771/50000 [13:22<00:18, 65.85it/s]




 98%|█████████▊| 48778/50000 [13:22<00:18, 65.50it/s]




 98%|█████████▊| 48785/50000 [13:22<00:19, 63.91it/s]




 98%|█████████▊| 48792/50000 [13:22<00:18, 63.88it/s]




 98%|█████████▊| 48799/50000 [13:22<00:19, 61.35it/s]




 98%|█████████▊| 48806/50000 [13:23<00:20, 57.42it/s]




 98%|█████████▊| 48813/50000 [13:23<00:20, 58.53it/s]




 98%|█████████▊| 48820/50000 [13:23<00:19, 60.91it/s]




 98%|█████████▊| 48827/50000 [13:23<00:18, 62.16it/s]




 98%|█████████▊| 48835/50000 [13:23<00:18, 64.64it/s]




 98%|█████████▊| 48842/50000 [13:23<00:17, 65.84it/s]




 98%|█████████▊| 48849/50000 [13:23<00:17, 63.97it/s]




 98%|█████████▊| 48856/50000 [13:23<00:18, 62.39it/s]




 98%|█████████▊| 48863/50000 [13:23<00:17, 63.83it/s]




 98%|█████████▊| 48870/50000 [13:24<00:18, 62.13it/s]




 98%|█████████▊| 48877/50000 [13:24<00:19, 58.11it/s]




 98%|█████████▊| 48884/50000 [13:24<00:18, 58.81it/s]




 98%|█████████▊| 48891/50000 [13:24<00:18, 60.46it/s]




 98%|█████████▊| 48898/50000 [13:24<00:17, 62.91it/s]




 98%|█████████▊| 48905/50000 [13:24<00:18, 60.47it/s]




 98%|█████████▊| 48913/50000 [13:24<00:17, 63.73it/s]




 98%|█████████▊| 48921/50000 [13:24<00:16, 67.41it/s]




 98%|█████████▊| 48928/50000 [13:24<00:16, 65.73it/s]




 98%|█████████▊| 48935/50000 [13:25<00:16, 64.07it/s]




 98%|█████████▊| 48942/50000 [13:25<00:17, 58.99it/s]




 98%|█████████▊| 48949/50000 [13:25<00:17, 58.54it/s]




 98%|█████████▊| 48955/50000 [13:25<00:19, 54.23it/s]




 98%|█████████▊| 48962/50000 [13:25<00:18, 55.97it/s]




 98%|█████████▊| 48969/50000 [13:25<00:17, 59.29it/s]




 98%|█████████▊| 48977/50000 [13:25<00:16, 63.26it/s]




 98%|█████████▊| 48984/50000 [13:25<00:16, 61.50it/s]




 98%|█████████▊| 48991/50000 [13:26<00:16, 60.68it/s]




 98%|█████████▊| 48998/50000 [13:26<00:16, 60.79it/s]




 98%|█████████▊| 49005/50000 [13:26<00:15, 62.65it/s]




 98%|█████████▊| 49012/50000 [13:26<00:15, 64.30it/s]




 98%|█████████▊| 49019/50000 [13:26<00:15, 65.00it/s]




 98%|█████████▊| 49026/50000 [13:26<00:15, 63.76it/s]




 98%|█████████▊| 49033/50000 [13:26<00:15, 62.58it/s]




 98%|█████████▊| 49040/50000 [13:26<00:15, 63.45it/s]




 98%|█████████▊| 49048/50000 [13:26<00:14, 66.03it/s]




 98%|█████████▊| 49055/50000 [13:27<00:14, 65.16it/s]




 98%|█████████▊| 49062/50000 [13:27<00:14, 66.22it/s]




 98%|█████████▊| 49069/50000 [13:27<00:14, 63.69it/s]




 98%|█████████▊| 49076/50000 [13:27<00:14, 62.04it/s]




 98%|█████████▊| 49084/50000 [13:27<00:14, 65.34it/s]




 98%|█████████▊| 49092/50000 [13:27<00:13, 67.27it/s]




 98%|█████████▊| 49100/50000 [13:27<00:13, 68.72it/s]




 98%|█████████▊| 49107/50000 [13:27<00:13, 65.84it/s]




 98%|█████████▊| 49114/50000 [13:27<00:13, 63.62it/s]




 98%|█████████▊| 49122/50000 [13:28<00:13, 66.98it/s]




 98%|█████████▊| 49129/50000 [13:28<00:13, 66.09it/s]




 98%|█████████▊| 49136/50000 [13:28<00:14, 60.80it/s]




 98%|█████████▊| 49143/50000 [13:28<00:14, 58.88it/s]




 98%|█████████▊| 49150/50000 [13:28<00:14, 59.36it/s]




 98%|█████████▊| 49157/50000 [13:28<00:14, 60.00it/s]




 98%|█████████▊| 49166/50000 [13:28<00:12, 64.84it/s]




 98%|█████████▊| 49173/50000 [13:28<00:12, 64.17it/s]




 98%|█████████▊| 49180/50000 [13:28<00:13, 62.27it/s]




 98%|█████████▊| 49187/50000 [13:29<00:13, 60.30it/s]




 98%|█████████▊| 49194/50000 [13:29<00:12, 62.79it/s]




 98%|█████████▊| 49201/50000 [13:29<00:12, 62.59it/s]




 98%|█████████▊| 49208/50000 [13:29<00:12, 63.42it/s]




 98%|█████████▊| 49215/50000 [13:29<00:12, 61.16it/s]




 98%|█████████▊| 49222/50000 [13:29<00:13, 59.49it/s]




 98%|█████████▊| 49229/50000 [13:29<00:12, 61.36it/s]




 98%|█████████▊| 49237/50000 [13:29<00:12, 63.21it/s]




 98%|█████████▊| 49244/50000 [13:29<00:11, 63.56it/s]




 99%|█████████▊| 49251/50000 [13:30<00:12, 61.46it/s]




 99%|█████████▊| 49258/50000 [13:30<00:12, 60.69it/s]




 99%|█████████▊| 49266/50000 [13:30<00:11, 63.91it/s]




 99%|█████████▊| 49273/50000 [13:30<00:11, 65.49it/s]




 99%|█████████▊| 49280/50000 [13:30<00:11, 60.76it/s]




 99%|█████████▊| 49287/50000 [13:30<00:11, 60.90it/s]




 99%|█████████▊| 49294/50000 [13:30<00:12, 58.79it/s]




 99%|█████████▊| 49300/50000 [13:30<00:12, 54.38it/s]




 99%|█████████▊| 49306/50000 [13:31<00:13, 51.20it/s]




 99%|█████████▊| 49313/50000 [13:31<00:12, 54.11it/s]




 99%|█████████▊| 49320/50000 [13:31<00:12, 56.15it/s]




 99%|█████████▊| 49327/50000 [13:31<00:11, 58.10it/s]




 99%|█████████▊| 49335/50000 [13:31<00:10, 61.31it/s]




 99%|█████████▊| 49342/50000 [13:31<00:10, 62.53it/s]




 99%|█████████▊| 49349/50000 [13:31<00:10, 62.57it/s]




 99%|█████████▊| 49356/50000 [13:31<00:11, 58.38it/s]




 99%|█████████▊| 49364/50000 [13:31<00:10, 62.68it/s]




 99%|█████████▊| 49371/50000 [13:32<00:10, 59.79it/s]




 99%|█████████▉| 49378/50000 [13:32<00:10, 58.21it/s]




 99%|█████████▉| 49384/50000 [13:32<00:10, 56.63it/s]




 99%|█████████▉| 49391/50000 [13:32<00:10, 58.89it/s]




 99%|█████████▉| 49397/50000 [13:32<00:11, 52.24it/s]




 99%|█████████▉| 49403/50000 [13:32<00:11, 52.06it/s]




 99%|█████████▉| 49412/50000 [13:32<00:10, 57.96it/s]




 99%|█████████▉| 49419/50000 [13:32<00:09, 59.00it/s]




 99%|█████████▉| 49426/50000 [13:33<00:09, 60.99it/s]




 99%|█████████▉| 49433/50000 [13:33<00:09, 61.48it/s]




 99%|█████████▉| 49440/50000 [13:33<00:09, 61.51it/s]




 99%|█████████▉| 49448/50000 [13:33<00:08, 64.54it/s]




 99%|█████████▉| 49455/50000 [13:33<00:09, 57.52it/s]




 99%|█████████▉| 49461/50000 [13:33<00:09, 55.54it/s]




 99%|█████████▉| 49467/50000 [13:33<00:10, 52.67it/s]




 99%|█████████▉| 49473/50000 [13:33<00:10, 51.22it/s]




 99%|█████████▉| 49479/50000 [13:34<00:11, 47.33it/s]




 99%|█████████▉| 49484/50000 [13:34<00:11, 44.25it/s]




 99%|█████████▉| 49489/50000 [13:34<00:12, 42.43it/s]




 99%|█████████▉| 49494/50000 [13:34<00:11, 44.07it/s]




 99%|█████████▉| 49499/50000 [13:34<00:12, 41.11it/s]




 99%|█████████▉| 49505/50000 [13:34<00:11, 44.32it/s]




 99%|█████████▉| 49510/50000 [13:34<00:11, 44.22it/s]




 99%|█████████▉| 49515/50000 [13:34<00:10, 44.15it/s]




 99%|█████████▉| 49520/50000 [13:35<00:12, 39.89it/s]




 99%|█████████▉| 49525/50000 [13:35<00:11, 42.38it/s]




 99%|█████████▉| 49532/50000 [13:35<00:10, 46.71it/s]




 99%|█████████▉| 49539/50000 [13:35<00:08, 51.35it/s]




 99%|█████████▉| 49546/50000 [13:35<00:08, 55.06it/s]




 99%|█████████▉| 49552/50000 [13:35<00:08, 55.86it/s]




 99%|█████████▉| 49558/50000 [13:35<00:07, 56.92it/s]




 99%|█████████▉| 49564/50000 [13:35<00:08, 54.33it/s]




 99%|█████████▉| 49570/50000 [13:35<00:08, 48.62it/s]




 99%|█████████▉| 49576/50000 [13:36<00:08, 50.29it/s]




 99%|█████████▉| 49582/50000 [13:36<00:09, 45.56it/s]




 99%|█████████▉| 49587/50000 [13:36<00:09, 41.75it/s]




 99%|█████████▉| 49593/50000 [13:36<00:09, 44.88it/s]




 99%|█████████▉| 49599/50000 [13:36<00:08, 47.66it/s]




 99%|█████████▉| 49606/50000 [13:36<00:07, 51.01it/s]




 99%|█████████▉| 49612/50000 [13:36<00:08, 46.62it/s]




 99%|█████████▉| 49617/50000 [13:36<00:08, 44.45it/s]




 99%|█████████▉| 49622/50000 [13:37<00:08, 42.83it/s]




 99%|█████████▉| 49628/50000 [13:37<00:08, 46.13it/s]




 99%|█████████▉| 49634/50000 [13:37<00:07, 47.77it/s]




 99%|█████████▉| 49641/50000 [13:37<00:06, 51.44it/s]




 99%|█████████▉| 49647/50000 [13:37<00:06, 53.49it/s]




 99%|█████████▉| 49654/50000 [13:37<00:06, 56.25it/s]




 99%|█████████▉| 49660/50000 [13:37<00:06, 50.44it/s]




 99%|█████████▉| 49666/50000 [13:37<00:06, 49.91it/s]




 99%|█████████▉| 49673/50000 [13:38<00:06, 53.11it/s]




 99%|█████████▉| 49680/50000 [13:38<00:05, 56.53it/s]




 99%|█████████▉| 49687/50000 [13:38<00:05, 58.53it/s]




 99%|█████████▉| 49693/50000 [13:38<00:05, 55.42it/s]




 99%|█████████▉| 49699/50000 [13:38<00:05, 55.51it/s]




 99%|█████████▉| 49706/50000 [13:38<00:04, 59.07it/s]




 99%|█████████▉| 49715/50000 [13:38<00:04, 65.61it/s]




 99%|█████████▉| 49722/50000 [13:38<00:04, 64.52it/s]




 99%|█████████▉| 49729/50000 [13:38<00:04, 64.84it/s]




 99%|█████████▉| 49736/50000 [13:39<00:04, 63.39it/s]




 99%|█████████▉| 49743/50000 [13:39<00:04, 62.66it/s]




100%|█████████▉| 49752/50000 [13:39<00:03, 67.44it/s]




100%|█████████▉| 49759/50000 [13:39<00:03, 67.85it/s]




100%|█████████▉| 49767/50000 [13:39<00:03, 70.39it/s]




100%|█████████▉| 49775/50000 [13:39<00:03, 63.07it/s]




100%|█████████▉| 49782/50000 [13:39<00:03, 64.24it/s]




100%|█████████▉| 49789/50000 [13:39<00:03, 61.58it/s]




100%|█████████▉| 49796/50000 [13:39<00:03, 59.84it/s]




100%|█████████▉| 49803/50000 [13:40<00:03, 59.89it/s]




100%|█████████▉| 49810/50000 [13:40<00:03, 58.57it/s]




100%|█████████▉| 49817/50000 [13:40<00:03, 60.99it/s]




100%|█████████▉| 49824/50000 [13:40<00:02, 61.64it/s]




100%|█████████▉| 49831/50000 [13:40<00:02, 59.43it/s]




100%|█████████▉| 49838/50000 [13:40<00:02, 60.44it/s]




100%|█████████▉| 49845/50000 [13:40<00:02, 60.07it/s]




100%|█████████▉| 49852/50000 [13:40<00:02, 62.12it/s]




100%|█████████▉| 49859/50000 [13:41<00:02, 61.46it/s]




100%|█████████▉| 49866/50000 [13:41<00:02, 60.38it/s]




100%|█████████▉| 49873/50000 [13:41<00:02, 56.28it/s]




100%|█████████▉| 49879/50000 [13:41<00:02, 55.29it/s]




100%|█████████▉| 49885/50000 [13:41<00:02, 55.64it/s]




100%|█████████▉| 49891/50000 [13:41<00:01, 55.58it/s]




100%|█████████▉| 49897/50000 [13:41<00:02, 51.28it/s]




100%|█████████▉| 49903/50000 [13:41<00:01, 52.67it/s]




100%|█████████▉| 49909/50000 [13:41<00:01, 54.27it/s]




100%|█████████▉| 49915/50000 [13:42<00:01, 51.20it/s]




100%|█████████▉| 49923/50000 [13:42<00:01, 55.52it/s]




100%|█████████▉| 49930/50000 [13:42<00:01, 58.10it/s]




100%|█████████▉| 49937/50000 [13:42<00:01, 59.70it/s]




100%|█████████▉| 49945/50000 [13:42<00:00, 63.58it/s]




100%|█████████▉| 49952/50000 [13:42<00:00, 58.26it/s]




100%|█████████▉| 49959/50000 [13:42<00:00, 58.48it/s]




100%|█████████▉| 49966/50000 [13:42<00:00, 60.60it/s]




100%|█████████▉| 49973/50000 [13:43<00:00, 59.67it/s]




100%|█████████▉| 49980/50000 [13:43<00:00, 59.53it/s]




100%|█████████▉| 49987/50000 [13:43<00:00, 59.52it/s]




100%|█████████▉| 49993/50000 [13:43<00:00, 59.36it/s]




100%|██████████| 50000/50000 [13:43<00:00, 60.72it/s]

data1["pos"] = pos
data1["neg"] = neg
data1["neu"] = neu
data1["compound"] = compound

Essays and title word count

essay_word_count = []
for ess in data1["preprocessed_essays"] :
    c = len(ess.split())
    essay_word_count.append(c) 

data1['essay_word_count'] = essay_word_count

title_word_count = []
for ess in data1["preprocessed_titles"] :
    c = len(ess.split())
    title_word_count.append(c) 

data1['title_word_count'] = title_word_count

y = data1['project_is_approved'].values
X = data1.drop(['project_is_approved'], axis=1)
X.head(1)

	school_state	teacher_number_of_previously_posted_projects	clean_categories	clean_subcategories	clean_teacher_prefix	clean_project_grade_category	price	quantity	preprocessed_essays	preprocessed_titles	pos	neg	neu	compound	essay_word_count	title_word_count
0	IN	0	Literacy_Language	ESL Literacy	Mrs	PreK-2	154.6	23	my students english learners working english s...	educational support english learners home	0.144	0.012	0.844	0.9694	161	5

# train test split
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, stratify=y)
X_train, X_cv, y_train, y_cv = train_test_split(X_train, y_train, test_size=0.33, stratify=y_train)

pos vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['pos'].values.reshape(1,-1))

X_train_pos_norm = normalizer.transform(X_train['pos'].values.reshape(1,-1))
X_cv_pos_norm = normalizer.transform(X_cv['pos'].values.reshape(1,-1))
X_test_pos_norm = normalizer.transform(X_test['pos'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_pos_norm.shape, y_train.shape)
print(X_cv_pos_norm.shape, y_cv.shape)
print(X_test_pos_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of pos")

X_train_pos_norm = X_train_pos_norm.transpose()
X_cv_pos_norm = X_cv_pos_norm.transpose()
X_test_pos_norm = X_test_pos_norm.transpose()


print("After vectorizations")
print(X_train_pos_norm.shape, y_train.shape)
print(X_cv_pos_norm.shape, y_cv.shape)
print(X_test_pos_norm.shape, y_test.shape)
print("="*100)

Transpose of pos
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

neg vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['neg'].values.reshape(1,-1))

X_train_neg_norm = normalizer.transform(X_train['neg'].values.reshape(1,-1))
X_cv_neg_norm = normalizer.transform(X_cv['neg'].values.reshape(1,-1))
X_test_neg_norm = normalizer.transform(X_test['neg'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_neg_norm.shape, y_train.shape)
print(X_cv_neg_norm.shape, y_cv.shape)
print(X_test_neg_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of neg")

X_train_neg_norm = X_train_neg_norm.transpose()
X_cv_neg_norm = X_cv_neg_norm.transpose()
X_test_neg_norm = X_test_neg_norm.transpose()


print("After vectorizations")
print(X_train_neg_norm.shape, y_train.shape)
print(X_cv_neg_norm.shape, y_cv.shape)
print(X_test_neg_norm.shape, y_test.shape)
print("="*100)

Transpose of neg
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

neutral vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['neu'].values.reshape(1,-1))

X_train_neu_norm = normalizer.transform(X_train['neu'].values.reshape(1,-1))
X_cv_neu_norm = normalizer.transform(X_cv['neu'].values.reshape(1,-1))
X_test_neu_norm = normalizer.transform(X_test['neu'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_neu_norm.shape, y_train.shape)
print(X_cv_neu_norm.shape, y_cv.shape)
print(X_test_neu_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of neutral")

X_train_neu_norm = X_train_neu_norm.transpose()
X_cv_neu_norm = X_cv_neu_norm.transpose()
X_test_neu_norm = X_test_neu_norm.transpose()


print("After vectorizations")
print(X_train_neu_norm.shape, y_train.shape)
print(X_cv_neu_norm.shape, y_cv.shape)
print(X_test_neu_norm.shape, y_test.shape)
print("="*100)

Transpose of neutral
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

compound vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['compound'].values.reshape(1,-1))

X_train_compound_norm = normalizer.transform(X_train['compound'].values.reshape(1,-1))
X_cv_compound_norm = normalizer.transform(X_cv['compound'].values.reshape(1,-1))
X_test_compound_norm = normalizer.transform(X_test['compound'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_compound_norm.shape, y_train.shape)
print(X_cv_compound_norm.shape, y_cv.shape)
print(X_test_compound_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of compound")

X_train_compound_norm = X_train_compound_norm.transpose()
X_cv_compound_norm = X_cv_compound_norm.transpose()
X_test_compound_norm = X_test_compound_norm.transpose()


print("After vectorizations")
print(X_train_compound_norm.shape, y_train.shape)
print(X_cv_compound_norm.shape, y_cv.shape)
print(X_test_compound_norm.shape, y_test.shape)
print("="*100)

Transpose of compound
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

essay word count vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['essay_word_count'].values.reshape(1,-1))

X_train_essay_word_count_norm = normalizer.transform(X_train['essay_word_count'].values.reshape(1,-1))
X_cv_essay_word_count_norm = normalizer.transform(X_cv['essay_word_count'].values.reshape(1,-1))
X_test_essay_word_count_norm = normalizer.transform(X_test['essay_word_count'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_essay_word_count_norm.shape, y_train.shape)
print(X_cv_essay_word_count_norm.shape, y_cv.shape)
print(X_test_essay_word_count_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of essay word count norm")

X_train_essay_word_count_norm = X_train_essay_word_count_norm.transpose()
X_cv_essay_word_count_norm = X_cv_essay_word_count_norm.transpose()
X_test_essay_word_count_norm = X_test_essay_word_count_norm.transpose()


print("After vectorizations")
print(X_train_essay_word_count_norm.shape, y_train.shape)
print(X_cv_essay_word_count_norm.shape, y_cv.shape)
print(X_test_essay_word_count_norm.shape, y_test.shape)
print("="*100)

Transpose of essay word count norm
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

title word count vectorization

from sklearn.preprocessing import Normalizer
normalizer = Normalizer()
# normalizer.fit(X_train['price'].values)
# this will rise an error Expected 2D array, got 1D array instead: 
# array=[105.22 215.96  96.01 ... 368.98  80.53 709.67].
# Reshape your data either using 
# array.reshape(-1, 1) if your data has a single feature 
# array.reshape(1, -1)  if it contains a single sample.
normalizer.fit(X_train['title_word_count'].values.reshape(1,-1))

X_train_title_word_count_norm = normalizer.transform(X_train['title_word_count'].values.reshape(1,-1))
X_cv_title_word_count_norm = normalizer.transform(X_cv['title_word_count'].values.reshape(1,-1))
X_test_title_word_count_norm = normalizer.transform(X_test['title_word_count'].values.reshape(1,-1))

print("After vectorizations")
print(X_train_title_word_count_norm.shape, y_train.shape)
print(X_cv_title_word_count_norm.shape, y_cv.shape)
print(X_test_title_word_count_norm.shape, y_test.shape)
print("="*100)

After vectorizations
(1, 22445) (22445,)
(1, 11055) (11055,)
(1, 16500) (16500,)
====================================================================================================

print("Transpose of essay word count norm")

X_train_title_word_count_norm = X_train_title_word_count_norm.transpose()
X_cv_title_word_count_norm = X_cv_title_word_count_norm.transpose()
X_test_title_word_count_norm = X_test_title_word_count_norm.transpose()


print("After vectorizations")
print(X_train_title_word_count_norm.shape, y_train.shape)
print(X_cv_title_word_count_norm.shape, y_cv.shape)
print(X_test_title_word_count_norm.shape, y_test.shape)
print("="*100)

Transpose of essay word count norm
After vectorizations
(22445, 1) (22445,)
(11055, 1) (11055,)
(16500, 1) (16500,)
====================================================================================================

Set 5

# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
X_tr_NO_TEXT = hstack((X_train_pos_norm, X_train_neg_norm, X_train_neu_norm, X_train_compound_norm, X_train_essay_word_count_norm, X_train_title_word_count_norm, X_train_state_ohe, X_train_teacher_ohe, X_train_grade_ohe, X_train_CSC_ohe, X_train_CC_ohe, X_train_price_norm, X_train_quantity_norm, X_train_TPPP_norm)).tocsr()
X_cr_NO_TEXT = hstack((X_cv_pos_norm, X_cv_neg_norm, X_cv_neu_norm, X_cv_compound_norm, X_cv_essay_word_count_norm, X_cv_title_word_count_norm, X_cv_state_ohe, X_cv_teacher_ohe, X_cv_grade_ohe, X_cv_CSC_ohe, X_cv_CC_ohe, X_cv_price_norm, X_cv_quantity_norm, X_cv_TPPP_norm)).tocsr()
X_te_NO_TEXT = hstack((X_test_pos_norm, X_test_neg_norm, X_test_neu_norm, X_test_compound_norm, X_test_essay_word_count_norm, X_test_title_word_count_norm, X_test_state_ohe, X_test_teacher_ohe, X_test_grade_ohe, X_test_CSC_ohe, X_test_CC_ohe, X_test_price_norm, X_test_quantity_norm, X_test_TPPP_norm)).tocsr()

print("Final Data matrix")
print(X_tr_NO_TEXT.shape, y_train.shape)
print(X_cr_NO_TEXT.shape, y_cv.shape)
print(X_te_NO_TEXT.shape, y_test.shape)
print("="*100)

Final Data matrix
(22445, 108) (22445,)
(11055, 108) (11055,)
(16500, 108) (16500,)
====================================================================================================

Logistic regression with no text data

import warnings
warnings.filterwarnings('ignore')
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
#from sklearn.grid_search import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import learning_curve, GridSearchCV

"""
y_true : array, shape = [n_samples] or [n_samples, n_classes]
True binary labels or binary label indicators.

y_score : array, shape = [n_samples] or [n_samples, n_classes]
Target scores, can either be probability estimates of the positive class, confidence values, or non-thresholded measure of
decisions (as returned by “decision_function” on some classifiers). 
For binary y_true, y_score is supposed to be the score of the class with greater label.

"""



clf = LogisticRegression(class_weight='balanced');
parameters ={'C':[10**-4, 10**-3,10**-2,1,10,100,1000,500,1000,10000]}
sd=GridSearchCV(clf, parameters, cv=5, scoring='roc_auc',return_train_score=True)
sd.fit(X_tr_NO_TEXT, y_train);


train_auc= sd.cv_results_['mean_train_score']
train_auc_std= sd.cv_results_['std_train_score']
cv_auc = sd.cv_results_['mean_test_score'] 
cv_auc_std= sd.cv_results_['std_test_score']

plt.plot(parameters['C'], train_auc, label='Train AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],train_auc - train_auc_std,train_auc + train_auc_std,alpha=0.2,color='darkblue')

plt.plot(parameters['C'], cv_auc, label='CV AUC')
# this code is copied from here: https://stackoverflow.com/a/48803361/4084039
plt.gca().fill_between(parameters['C'],cv_auc - cv_auc_std,cv_auc + cv_auc_std,alpha=0.2,color='darkorange')

plt.scatter(parameters['C'], train_auc, label='Train AUC points')
plt.scatter(parameters['C'], cv_auc, label='CV AUC points')
plt.xscale('log') 

plt.legend()
plt.xlabel("C(1/lambda): hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

##Fitting Model to Hyper-Parameter Curve
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc


neigh = LogisticRegression(C=10**3,class_weight='balanced');
neigh.fit(X_tr_NO_TEXT ,y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

train_fpr, train_tpr, thresholds = roc_curve(y_train, neigh.predict_proba(X_tr_NO_TEXT)[:,1])
test_fpr, test_tpr, thresholds = roc_curve(y_test, neigh.predict_proba(X_te_NO_TEXT)[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.ylabel("True Positive Rate(TPR)")
plt.xlabel("False Positive Rate(FPR)")
plt.title("ROC PLOTS")
plt.show()

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_train, neigh.predict(X_tr_NO_TEXT )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

#https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
import seaborn as sns
import matplotlib.pyplot as plt     

ax= plt.subplot()
sns.heatmap(confusion_matrix(y_test, neigh.predict(X_te_NO_TEXT )), annot=True, ax = ax,fmt='g'); #annot=True to annotate cells

# labels, title and ticks
ax.set_xlabel('Predicted labels');
ax.set_ylabel('True labels'); 
ax.set_title('Confusion Matrix'); 
#ax.xaxis.set_ticklabels(['business', 'health']); ax.yaxis.set_ticklabels(['health', 'business']);

3. Conclusion

# Please compare all your models using Prettytable library

# Please compare all your models using Prettytable library
# http://zetcode.com/python/prettytable/

from prettytable import PrettyTable

#If you get a ModuleNotFoundError error , install prettytable using: pip3 install prettytable

x = PrettyTable()
x.field_names = ["Vectorizer", "Model", "Alpha:Hyper Parameter", "AUC"]

x.add_row(["BOW", "Logistic Regression", 0.01, 0.65])
x.add_row(["TFIDF", "Logistic Regression", 1, 0.64])
x.add_row(["AVG W2V", "Logistic Regression", 1, 0.68])
x.add_row(["TFIDF W2V", "Logistic Regression", 0.01, 0.68])
x.add_row(["NO TEXT", "Logistic Regression", 1000, 0.53])


print(x)

+------------+---------------------+-----------------------+------+
| Vectorizer |        Model        | Alpha:Hyper Parameter | AUC  |
+------------+---------------------+-----------------------+------+
|    BOW     | Logistic Regression |          0.01         | 0.65 |
|   TFIDF    | Logistic Regression |           1           | 0.64 |
|  AVG W2V   | Logistic Regression |           1           | 0.68 |
| TFIDF W2V  | Logistic Regression |          0.01         | 0.68 |
|  NO TEXT   | Logistic Regression |          1000         | 0.53 |
+------------+---------------------+-----------------------+------+

Observation

1. Without text data there is a noticable difference observed in AUC score.
2. TFIDF W2V and AVG W2V has given same AUC score with different hyper parameters.